Upload folder using huggingface_hub
Browse files- checkpoint-130/config.json +64 -0
- checkpoint-130/model.safetensors +3 -0
- checkpoint-130/optimizer.pt +3 -0
- checkpoint-130/preprocessor_config.json +36 -0
- checkpoint-130/rng_state.pth +3 -0
- checkpoint-130/scheduler.pt +3 -0
- checkpoint-130/trainer_state.json +1048 -0
- checkpoint-130/training_args.bin +3 -0
- checkpoint-190/config.json +64 -0
- checkpoint-190/model.safetensors +3 -0
- checkpoint-190/optimizer.pt +3 -0
- checkpoint-190/preprocessor_config.json +36 -0
- checkpoint-190/rng_state.pth +3 -0
- checkpoint-190/scheduler.pt +3 -0
- checkpoint-190/trainer_state.json +1522 -0
- checkpoint-190/training_args.bin +3 -0
- checkpoint-200/config.json +64 -0
- checkpoint-200/model.safetensors +3 -0
- checkpoint-200/optimizer.pt +3 -0
- checkpoint-200/preprocessor_config.json +36 -0
- checkpoint-200/rng_state.pth +3 -0
- checkpoint-200/scheduler.pt +3 -0
- checkpoint-200/trainer_state.json +1601 -0
- checkpoint-200/training_args.bin +3 -0
- config.json +64 -0
- model.safetensors +3 -0
- preprocessor_config.json +36 -0
- training_args.bin +3 -0
checkpoint-130/config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Organika/sdxl-detector",
|
| 3 |
+
"_num_labels": 2,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"SwinForImageClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.0,
|
| 8 |
+
"depths": [
|
| 9 |
+
2,
|
| 10 |
+
2,
|
| 11 |
+
18,
|
| 12 |
+
2
|
| 13 |
+
],
|
| 14 |
+
"drop_path_rate": 0.1,
|
| 15 |
+
"embed_dim": 128,
|
| 16 |
+
"encoder_stride": 32,
|
| 17 |
+
"hidden_act": "gelu",
|
| 18 |
+
"hidden_dropout_prob": 0.0,
|
| 19 |
+
"hidden_size": 1024,
|
| 20 |
+
"id2label": {
|
| 21 |
+
"0": "artificial",
|
| 22 |
+
"1": "human"
|
| 23 |
+
},
|
| 24 |
+
"image_size": 224,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"label2id": {
|
| 27 |
+
"artificial": 0,
|
| 28 |
+
"human": 1
|
| 29 |
+
},
|
| 30 |
+
"layer_norm_eps": 1e-05,
|
| 31 |
+
"max_length": 128,
|
| 32 |
+
"mlp_ratio": 4.0,
|
| 33 |
+
"model_type": "swin",
|
| 34 |
+
"num_channels": 3,
|
| 35 |
+
"num_heads": [
|
| 36 |
+
4,
|
| 37 |
+
8,
|
| 38 |
+
16,
|
| 39 |
+
32
|
| 40 |
+
],
|
| 41 |
+
"num_layers": 4,
|
| 42 |
+
"out_features": [
|
| 43 |
+
"stage4"
|
| 44 |
+
],
|
| 45 |
+
"out_indices": [
|
| 46 |
+
4
|
| 47 |
+
],
|
| 48 |
+
"padding": "max_length",
|
| 49 |
+
"patch_size": 4,
|
| 50 |
+
"path_norm": true,
|
| 51 |
+
"problem_type": "single_label_classification",
|
| 52 |
+
"qkv_bias": true,
|
| 53 |
+
"stage_names": [
|
| 54 |
+
"stem",
|
| 55 |
+
"stage1",
|
| 56 |
+
"stage2",
|
| 57 |
+
"stage3",
|
| 58 |
+
"stage4"
|
| 59 |
+
],
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.39.3",
|
| 62 |
+
"use_absolute_embeddings": false,
|
| 63 |
+
"window_size": 7
|
| 64 |
+
}
|
checkpoint-130/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86a9c215382ea7372ff3d6cc70c3c6268b24b661c627fbfe8ef12ac3abd2a13a
|
| 3 |
+
size 347498816
|
checkpoint-130/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d7a56e1fa8ec2cfa10deda0b0bfd4d77ee75bf9fd7acee7ec54c35e82a6c0b6
|
| 3 |
+
size 694317645
|
checkpoint-130/preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.485,
|
| 21 |
+
0.456,
|
| 22 |
+
0.406
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.229,
|
| 27 |
+
0.224,
|
| 28 |
+
0.225
|
| 29 |
+
],
|
| 30 |
+
"resample": 3,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
checkpoint-130/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e6dd85f936f43466f39788b2b1110fbce547a1701a01c573f3d90de30d587fe
|
| 3 |
+
size 14244
|
checkpoint-130/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d057eafda7411f97093737064830a708851e776787dd239516be91eca678ee5d
|
| 3 |
+
size 1064
|
checkpoint-130/trainer_state.json
ADDED
|
@@ -0,0 +1,1048 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.004848334938287735,
|
| 3 |
+
"best_model_checkpoint": "ai_vs_real_image_detection/checkpoint-130",
|
| 4 |
+
"epoch": 2.6,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 130,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 52.89345169067383,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 2.2763,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04,
|
| 20 |
+
"grad_norm": 62.66461944580078,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.8254,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06,
|
| 27 |
+
"grad_norm": 66.1015625,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 1.8906,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08,
|
| 34 |
+
"grad_norm": 56.700233459472656,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 1.3529,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1,
|
| 41 |
+
"grad_norm": 66.10917663574219,
|
| 42 |
+
"learning_rate": 2.5e-05,
|
| 43 |
+
"loss": 1.4672,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12,
|
| 48 |
+
"grad_norm": 52.68284225463867,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 1.0578,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14,
|
| 55 |
+
"grad_norm": 42.4580078125,
|
| 56 |
+
"learning_rate": 3.5e-05,
|
| 57 |
+
"loss": 1.1221,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.16,
|
| 62 |
+
"grad_norm": 64.87982940673828,
|
| 63 |
+
"learning_rate": 4e-05,
|
| 64 |
+
"loss": 0.9313,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.18,
|
| 69 |
+
"grad_norm": 56.227081298828125,
|
| 70 |
+
"learning_rate": 4.5e-05,
|
| 71 |
+
"loss": 0.644,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2,
|
| 76 |
+
"grad_norm": 13.719300270080566,
|
| 77 |
+
"learning_rate": 5e-05,
|
| 78 |
+
"loss": 0.4305,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2,
|
| 83 |
+
"eval_accuracy": 0.9144736842105263,
|
| 84 |
+
"eval_loss": 0.19980163872241974,
|
| 85 |
+
"eval_runtime": 6.726,
|
| 86 |
+
"eval_samples_per_second": 45.198,
|
| 87 |
+
"eval_steps_per_second": 1.189,
|
| 88 |
+
"step": 10
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.22,
|
| 92 |
+
"grad_norm": 16.044822692871094,
|
| 93 |
+
"learning_rate": 4.973684210526316e-05,
|
| 94 |
+
"loss": 0.3892,
|
| 95 |
+
"step": 11
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.24,
|
| 99 |
+
"grad_norm": 15.53343391418457,
|
| 100 |
+
"learning_rate": 4.9473684210526315e-05,
|
| 101 |
+
"loss": 0.3683,
|
| 102 |
+
"step": 12
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.26,
|
| 106 |
+
"grad_norm": 13.435264587402344,
|
| 107 |
+
"learning_rate": 4.921052631578947e-05,
|
| 108 |
+
"loss": 0.3594,
|
| 109 |
+
"step": 13
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.28,
|
| 113 |
+
"grad_norm": 17.45833969116211,
|
| 114 |
+
"learning_rate": 4.8947368421052635e-05,
|
| 115 |
+
"loss": 0.5304,
|
| 116 |
+
"step": 14
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.3,
|
| 120 |
+
"grad_norm": 14.26673412322998,
|
| 121 |
+
"learning_rate": 4.868421052631579e-05,
|
| 122 |
+
"loss": 0.4804,
|
| 123 |
+
"step": 15
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.32,
|
| 127 |
+
"grad_norm": 9.26322078704834,
|
| 128 |
+
"learning_rate": 4.842105263157895e-05,
|
| 129 |
+
"loss": 0.3633,
|
| 130 |
+
"step": 16
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.34,
|
| 134 |
+
"grad_norm": 8.355168342590332,
|
| 135 |
+
"learning_rate": 4.8157894736842105e-05,
|
| 136 |
+
"loss": 0.4711,
|
| 137 |
+
"step": 17
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.36,
|
| 141 |
+
"grad_norm": 8.733236312866211,
|
| 142 |
+
"learning_rate": 4.789473684210526e-05,
|
| 143 |
+
"loss": 0.3954,
|
| 144 |
+
"step": 18
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.38,
|
| 148 |
+
"grad_norm": 5.385647296905518,
|
| 149 |
+
"learning_rate": 4.7631578947368424e-05,
|
| 150 |
+
"loss": 0.1943,
|
| 151 |
+
"step": 19
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.4,
|
| 155 |
+
"grad_norm": 9.892962455749512,
|
| 156 |
+
"learning_rate": 4.736842105263158e-05,
|
| 157 |
+
"loss": 0.3209,
|
| 158 |
+
"step": 20
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.4,
|
| 162 |
+
"eval_accuracy": 0.8881578947368421,
|
| 163 |
+
"eval_loss": 0.24087880551815033,
|
| 164 |
+
"eval_runtime": 6.8213,
|
| 165 |
+
"eval_samples_per_second": 44.566,
|
| 166 |
+
"eval_steps_per_second": 1.173,
|
| 167 |
+
"step": 20
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.42,
|
| 171 |
+
"grad_norm": 5.46142578125,
|
| 172 |
+
"learning_rate": 4.7105263157894744e-05,
|
| 173 |
+
"loss": 0.1505,
|
| 174 |
+
"step": 21
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.44,
|
| 178 |
+
"grad_norm": 9.863460540771484,
|
| 179 |
+
"learning_rate": 4.68421052631579e-05,
|
| 180 |
+
"loss": 0.2481,
|
| 181 |
+
"step": 22
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.46,
|
| 185 |
+
"grad_norm": 7.595032691955566,
|
| 186 |
+
"learning_rate": 4.657894736842106e-05,
|
| 187 |
+
"loss": 0.2001,
|
| 188 |
+
"step": 23
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.48,
|
| 192 |
+
"grad_norm": 17.375865936279297,
|
| 193 |
+
"learning_rate": 4.6315789473684214e-05,
|
| 194 |
+
"loss": 0.336,
|
| 195 |
+
"step": 24
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.5,
|
| 199 |
+
"grad_norm": 5.597734451293945,
|
| 200 |
+
"learning_rate": 4.605263157894737e-05,
|
| 201 |
+
"loss": 0.1395,
|
| 202 |
+
"step": 25
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.52,
|
| 206 |
+
"grad_norm": 7.56685209274292,
|
| 207 |
+
"learning_rate": 4.5789473684210527e-05,
|
| 208 |
+
"loss": 0.1739,
|
| 209 |
+
"step": 26
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.54,
|
| 213 |
+
"grad_norm": 13.769929885864258,
|
| 214 |
+
"learning_rate": 4.552631578947369e-05,
|
| 215 |
+
"loss": 0.1959,
|
| 216 |
+
"step": 27
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.56,
|
| 220 |
+
"grad_norm": 10.677892684936523,
|
| 221 |
+
"learning_rate": 4.5263157894736846e-05,
|
| 222 |
+
"loss": 0.2015,
|
| 223 |
+
"step": 28
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.58,
|
| 227 |
+
"grad_norm": 3.645374059677124,
|
| 228 |
+
"learning_rate": 4.5e-05,
|
| 229 |
+
"loss": 0.0576,
|
| 230 |
+
"step": 29
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.6,
|
| 234 |
+
"grad_norm": 12.538244247436523,
|
| 235 |
+
"learning_rate": 4.473684210526316e-05,
|
| 236 |
+
"loss": 0.324,
|
| 237 |
+
"step": 30
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.6,
|
| 241 |
+
"eval_accuracy": 0.9802631578947368,
|
| 242 |
+
"eval_loss": 0.056003469973802567,
|
| 243 |
+
"eval_runtime": 6.4198,
|
| 244 |
+
"eval_samples_per_second": 47.353,
|
| 245 |
+
"eval_steps_per_second": 1.246,
|
| 246 |
+
"step": 30
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.62,
|
| 250 |
+
"grad_norm": 8.700516700744629,
|
| 251 |
+
"learning_rate": 4.4473684210526316e-05,
|
| 252 |
+
"loss": 0.1199,
|
| 253 |
+
"step": 31
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.64,
|
| 257 |
+
"grad_norm": 6.926602840423584,
|
| 258 |
+
"learning_rate": 4.421052631578947e-05,
|
| 259 |
+
"loss": 0.1471,
|
| 260 |
+
"step": 32
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.66,
|
| 264 |
+
"grad_norm": 8.461554527282715,
|
| 265 |
+
"learning_rate": 4.394736842105263e-05,
|
| 266 |
+
"loss": 0.0649,
|
| 267 |
+
"step": 33
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.68,
|
| 271 |
+
"grad_norm": 7.129871845245361,
|
| 272 |
+
"learning_rate": 4.368421052631579e-05,
|
| 273 |
+
"loss": 0.1167,
|
| 274 |
+
"step": 34
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.7,
|
| 278 |
+
"grad_norm": 20.945388793945312,
|
| 279 |
+
"learning_rate": 4.342105263157895e-05,
|
| 280 |
+
"loss": 0.3279,
|
| 281 |
+
"step": 35
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.72,
|
| 285 |
+
"grad_norm": 4.466580390930176,
|
| 286 |
+
"learning_rate": 4.3157894736842105e-05,
|
| 287 |
+
"loss": 0.0517,
|
| 288 |
+
"step": 36
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.74,
|
| 292 |
+
"grad_norm": 9.97375202178955,
|
| 293 |
+
"learning_rate": 4.289473684210527e-05,
|
| 294 |
+
"loss": 0.1241,
|
| 295 |
+
"step": 37
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.76,
|
| 299 |
+
"grad_norm": 6.459863662719727,
|
| 300 |
+
"learning_rate": 4.2631578947368425e-05,
|
| 301 |
+
"loss": 0.0859,
|
| 302 |
+
"step": 38
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.78,
|
| 306 |
+
"grad_norm": 11.704880714416504,
|
| 307 |
+
"learning_rate": 4.236842105263158e-05,
|
| 308 |
+
"loss": 0.1252,
|
| 309 |
+
"step": 39
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.8,
|
| 313 |
+
"grad_norm": 7.90261697769165,
|
| 314 |
+
"learning_rate": 4.210526315789474e-05,
|
| 315 |
+
"loss": 0.0541,
|
| 316 |
+
"step": 40
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.8,
|
| 320 |
+
"eval_accuracy": 0.9671052631578947,
|
| 321 |
+
"eval_loss": 0.06965293735265732,
|
| 322 |
+
"eval_runtime": 6.6281,
|
| 323 |
+
"eval_samples_per_second": 45.865,
|
| 324 |
+
"eval_steps_per_second": 1.207,
|
| 325 |
+
"step": 40
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.82,
|
| 329 |
+
"grad_norm": 2.8429932594299316,
|
| 330 |
+
"learning_rate": 4.18421052631579e-05,
|
| 331 |
+
"loss": 0.0325,
|
| 332 |
+
"step": 41
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.84,
|
| 336 |
+
"grad_norm": 10.533605575561523,
|
| 337 |
+
"learning_rate": 4.157894736842106e-05,
|
| 338 |
+
"loss": 0.1836,
|
| 339 |
+
"step": 42
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.86,
|
| 343 |
+
"grad_norm": 12.08105754852295,
|
| 344 |
+
"learning_rate": 4.1315789473684214e-05,
|
| 345 |
+
"loss": 0.224,
|
| 346 |
+
"step": 43
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.88,
|
| 350 |
+
"grad_norm": 9.006488800048828,
|
| 351 |
+
"learning_rate": 4.105263157894737e-05,
|
| 352 |
+
"loss": 0.2139,
|
| 353 |
+
"step": 44
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9,
|
| 357 |
+
"grad_norm": 7.3875956535339355,
|
| 358 |
+
"learning_rate": 4.078947368421053e-05,
|
| 359 |
+
"loss": 0.088,
|
| 360 |
+
"step": 45
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.92,
|
| 364 |
+
"grad_norm": 13.019746780395508,
|
| 365 |
+
"learning_rate": 4.0526315789473684e-05,
|
| 366 |
+
"loss": 0.2363,
|
| 367 |
+
"step": 46
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.94,
|
| 371 |
+
"grad_norm": 37.95082473754883,
|
| 372 |
+
"learning_rate": 4.026315789473684e-05,
|
| 373 |
+
"loss": 0.3728,
|
| 374 |
+
"step": 47
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.96,
|
| 378 |
+
"grad_norm": 13.99156665802002,
|
| 379 |
+
"learning_rate": 4e-05,
|
| 380 |
+
"loss": 0.1376,
|
| 381 |
+
"step": 48
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.98,
|
| 385 |
+
"grad_norm": 5.063250541687012,
|
| 386 |
+
"learning_rate": 3.973684210526316e-05,
|
| 387 |
+
"loss": 0.0832,
|
| 388 |
+
"step": 49
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"grad_norm": 13.046541213989258,
|
| 393 |
+
"learning_rate": 3.9473684210526316e-05,
|
| 394 |
+
"loss": 0.1096,
|
| 395 |
+
"step": 50
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.0,
|
| 399 |
+
"eval_accuracy": 0.9572368421052632,
|
| 400 |
+
"eval_loss": 0.12372467666864395,
|
| 401 |
+
"eval_runtime": 6.8765,
|
| 402 |
+
"eval_samples_per_second": 44.209,
|
| 403 |
+
"eval_steps_per_second": 1.163,
|
| 404 |
+
"step": 50
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.02,
|
| 408 |
+
"grad_norm": 4.865447998046875,
|
| 409 |
+
"learning_rate": 3.921052631578947e-05,
|
| 410 |
+
"loss": 0.0366,
|
| 411 |
+
"step": 51
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.04,
|
| 415 |
+
"grad_norm": 12.430005073547363,
|
| 416 |
+
"learning_rate": 3.894736842105263e-05,
|
| 417 |
+
"loss": 0.1301,
|
| 418 |
+
"step": 52
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 1.06,
|
| 422 |
+
"grad_norm": 2.1291539669036865,
|
| 423 |
+
"learning_rate": 3.868421052631579e-05,
|
| 424 |
+
"loss": 0.0168,
|
| 425 |
+
"step": 53
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.08,
|
| 429 |
+
"grad_norm": 1.6922448873519897,
|
| 430 |
+
"learning_rate": 3.842105263157895e-05,
|
| 431 |
+
"loss": 0.0074,
|
| 432 |
+
"step": 54
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.1,
|
| 436 |
+
"grad_norm": 5.851077556610107,
|
| 437 |
+
"learning_rate": 3.815789473684211e-05,
|
| 438 |
+
"loss": 0.0216,
|
| 439 |
+
"step": 55
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.12,
|
| 443 |
+
"grad_norm": 21.35593605041504,
|
| 444 |
+
"learning_rate": 3.789473684210527e-05,
|
| 445 |
+
"loss": 0.1171,
|
| 446 |
+
"step": 56
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.14,
|
| 450 |
+
"grad_norm": 9.791412353515625,
|
| 451 |
+
"learning_rate": 3.7631578947368425e-05,
|
| 452 |
+
"loss": 0.1688,
|
| 453 |
+
"step": 57
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.16,
|
| 457 |
+
"grad_norm": 3.2542812824249268,
|
| 458 |
+
"learning_rate": 3.736842105263158e-05,
|
| 459 |
+
"loss": 0.0236,
|
| 460 |
+
"step": 58
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.18,
|
| 464 |
+
"grad_norm": 9.557928085327148,
|
| 465 |
+
"learning_rate": 3.710526315789474e-05,
|
| 466 |
+
"loss": 0.1782,
|
| 467 |
+
"step": 59
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.2,
|
| 471 |
+
"grad_norm": 12.514122009277344,
|
| 472 |
+
"learning_rate": 3.6842105263157895e-05,
|
| 473 |
+
"loss": 0.0905,
|
| 474 |
+
"step": 60
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.2,
|
| 478 |
+
"eval_accuracy": 0.9967105263157895,
|
| 479 |
+
"eval_loss": 0.027645185589790344,
|
| 480 |
+
"eval_runtime": 7.1325,
|
| 481 |
+
"eval_samples_per_second": 42.622,
|
| 482 |
+
"eval_steps_per_second": 1.122,
|
| 483 |
+
"step": 60
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.22,
|
| 487 |
+
"grad_norm": 4.434255599975586,
|
| 488 |
+
"learning_rate": 3.657894736842106e-05,
|
| 489 |
+
"loss": 0.0276,
|
| 490 |
+
"step": 61
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.24,
|
| 494 |
+
"grad_norm": 6.304673671722412,
|
| 495 |
+
"learning_rate": 3.6315789473684214e-05,
|
| 496 |
+
"loss": 0.0508,
|
| 497 |
+
"step": 62
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.26,
|
| 501 |
+
"grad_norm": 3.844200372695923,
|
| 502 |
+
"learning_rate": 3.605263157894737e-05,
|
| 503 |
+
"loss": 0.0199,
|
| 504 |
+
"step": 63
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.28,
|
| 508 |
+
"grad_norm": 0.8185235261917114,
|
| 509 |
+
"learning_rate": 3.578947368421053e-05,
|
| 510 |
+
"loss": 0.0069,
|
| 511 |
+
"step": 64
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.3,
|
| 515 |
+
"grad_norm": 6.909903049468994,
|
| 516 |
+
"learning_rate": 3.5526315789473684e-05,
|
| 517 |
+
"loss": 0.0597,
|
| 518 |
+
"step": 65
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.32,
|
| 522 |
+
"grad_norm": 3.5652379989624023,
|
| 523 |
+
"learning_rate": 3.526315789473684e-05,
|
| 524 |
+
"loss": 0.0141,
|
| 525 |
+
"step": 66
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.34,
|
| 529 |
+
"grad_norm": 11.72857666015625,
|
| 530 |
+
"learning_rate": 3.5e-05,
|
| 531 |
+
"loss": 0.1865,
|
| 532 |
+
"step": 67
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.36,
|
| 536 |
+
"grad_norm": 10.541884422302246,
|
| 537 |
+
"learning_rate": 3.473684210526316e-05,
|
| 538 |
+
"loss": 0.0606,
|
| 539 |
+
"step": 68
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.38,
|
| 543 |
+
"grad_norm": 5.210526943206787,
|
| 544 |
+
"learning_rate": 3.447368421052632e-05,
|
| 545 |
+
"loss": 0.0276,
|
| 546 |
+
"step": 69
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.4,
|
| 550 |
+
"grad_norm": 1.0411546230316162,
|
| 551 |
+
"learning_rate": 3.421052631578947e-05,
|
| 552 |
+
"loss": 0.009,
|
| 553 |
+
"step": 70
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.4,
|
| 557 |
+
"eval_accuracy": 0.9901315789473685,
|
| 558 |
+
"eval_loss": 0.020390018820762634,
|
| 559 |
+
"eval_runtime": 7.2971,
|
| 560 |
+
"eval_samples_per_second": 41.66,
|
| 561 |
+
"eval_steps_per_second": 1.096,
|
| 562 |
+
"step": 70
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.42,
|
| 566 |
+
"grad_norm": 0.7153753638267517,
|
| 567 |
+
"learning_rate": 3.3947368421052636e-05,
|
| 568 |
+
"loss": 0.0049,
|
| 569 |
+
"step": 71
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.44,
|
| 573 |
+
"grad_norm": 4.828935623168945,
|
| 574 |
+
"learning_rate": 3.368421052631579e-05,
|
| 575 |
+
"loss": 0.0297,
|
| 576 |
+
"step": 72
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.46,
|
| 580 |
+
"grad_norm": 9.103313446044922,
|
| 581 |
+
"learning_rate": 3.342105263157895e-05,
|
| 582 |
+
"loss": 0.071,
|
| 583 |
+
"step": 73
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.48,
|
| 587 |
+
"grad_norm": 7.635372161865234,
|
| 588 |
+
"learning_rate": 3.3157894736842106e-05,
|
| 589 |
+
"loss": 0.0363,
|
| 590 |
+
"step": 74
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.5,
|
| 594 |
+
"grad_norm": 8.57288646697998,
|
| 595 |
+
"learning_rate": 3.289473684210527e-05,
|
| 596 |
+
"loss": 0.0566,
|
| 597 |
+
"step": 75
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.52,
|
| 601 |
+
"grad_norm": 8.89769172668457,
|
| 602 |
+
"learning_rate": 3.2631578947368426e-05,
|
| 603 |
+
"loss": 0.0213,
|
| 604 |
+
"step": 76
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.54,
|
| 608 |
+
"grad_norm": 2.804738998413086,
|
| 609 |
+
"learning_rate": 3.236842105263158e-05,
|
| 610 |
+
"loss": 0.0138,
|
| 611 |
+
"step": 77
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.56,
|
| 615 |
+
"grad_norm": 4.841202259063721,
|
| 616 |
+
"learning_rate": 3.210526315789474e-05,
|
| 617 |
+
"loss": 0.0181,
|
| 618 |
+
"step": 78
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.58,
|
| 622 |
+
"grad_norm": 7.109033584594727,
|
| 623 |
+
"learning_rate": 3.1842105263157895e-05,
|
| 624 |
+
"loss": 0.0332,
|
| 625 |
+
"step": 79
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.6,
|
| 629 |
+
"grad_norm": 13.540718078613281,
|
| 630 |
+
"learning_rate": 3.157894736842105e-05,
|
| 631 |
+
"loss": 0.1667,
|
| 632 |
+
"step": 80
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.6,
|
| 636 |
+
"eval_accuracy": 0.9703947368421053,
|
| 637 |
+
"eval_loss": 0.10031529515981674,
|
| 638 |
+
"eval_runtime": 6.7491,
|
| 639 |
+
"eval_samples_per_second": 45.043,
|
| 640 |
+
"eval_steps_per_second": 1.185,
|
| 641 |
+
"step": 80
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.62,
|
| 645 |
+
"grad_norm": 30.279451370239258,
|
| 646 |
+
"learning_rate": 3.131578947368421e-05,
|
| 647 |
+
"loss": 0.1139,
|
| 648 |
+
"step": 81
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.64,
|
| 652 |
+
"grad_norm": 7.229616165161133,
|
| 653 |
+
"learning_rate": 3.105263157894737e-05,
|
| 654 |
+
"loss": 0.0284,
|
| 655 |
+
"step": 82
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.66,
|
| 659 |
+
"grad_norm": 0.327736496925354,
|
| 660 |
+
"learning_rate": 3.078947368421053e-05,
|
| 661 |
+
"loss": 0.0032,
|
| 662 |
+
"step": 83
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.68,
|
| 666 |
+
"grad_norm": 0.15309709310531616,
|
| 667 |
+
"learning_rate": 3.0526315789473684e-05,
|
| 668 |
+
"loss": 0.0011,
|
| 669 |
+
"step": 84
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.7,
|
| 673 |
+
"grad_norm": 13.175533294677734,
|
| 674 |
+
"learning_rate": 3.0263157894736844e-05,
|
| 675 |
+
"loss": 0.039,
|
| 676 |
+
"step": 85
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.72,
|
| 680 |
+
"grad_norm": 24.578662872314453,
|
| 681 |
+
"learning_rate": 3e-05,
|
| 682 |
+
"loss": 0.1834,
|
| 683 |
+
"step": 86
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.74,
|
| 687 |
+
"grad_norm": 6.690978527069092,
|
| 688 |
+
"learning_rate": 2.9736842105263157e-05,
|
| 689 |
+
"loss": 0.0214,
|
| 690 |
+
"step": 87
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.76,
|
| 694 |
+
"grad_norm": 21.097963333129883,
|
| 695 |
+
"learning_rate": 2.9473684210526314e-05,
|
| 696 |
+
"loss": 0.1569,
|
| 697 |
+
"step": 88
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.78,
|
| 701 |
+
"grad_norm": 7.984706401824951,
|
| 702 |
+
"learning_rate": 2.9210526315789477e-05,
|
| 703 |
+
"loss": 0.0326,
|
| 704 |
+
"step": 89
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.8,
|
| 708 |
+
"grad_norm": 11.008134841918945,
|
| 709 |
+
"learning_rate": 2.8947368421052634e-05,
|
| 710 |
+
"loss": 0.0602,
|
| 711 |
+
"step": 90
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.8,
|
| 715 |
+
"eval_accuracy": 0.9868421052631579,
|
| 716 |
+
"eval_loss": 0.052807241678237915,
|
| 717 |
+
"eval_runtime": 7.1111,
|
| 718 |
+
"eval_samples_per_second": 42.75,
|
| 719 |
+
"eval_steps_per_second": 1.125,
|
| 720 |
+
"step": 90
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.82,
|
| 724 |
+
"grad_norm": 21.149032592773438,
|
| 725 |
+
"learning_rate": 2.868421052631579e-05,
|
| 726 |
+
"loss": 0.0265,
|
| 727 |
+
"step": 91
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.84,
|
| 731 |
+
"grad_norm": 3.4383814334869385,
|
| 732 |
+
"learning_rate": 2.842105263157895e-05,
|
| 733 |
+
"loss": 0.0059,
|
| 734 |
+
"step": 92
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.86,
|
| 738 |
+
"grad_norm": 6.291805267333984,
|
| 739 |
+
"learning_rate": 2.8157894736842106e-05,
|
| 740 |
+
"loss": 0.0177,
|
| 741 |
+
"step": 93
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 1.88,
|
| 745 |
+
"grad_norm": 1.5197041034698486,
|
| 746 |
+
"learning_rate": 2.7894736842105263e-05,
|
| 747 |
+
"loss": 0.0065,
|
| 748 |
+
"step": 94
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 1.9,
|
| 752 |
+
"grad_norm": 0.47119140625,
|
| 753 |
+
"learning_rate": 2.7631578947368426e-05,
|
| 754 |
+
"loss": 0.0018,
|
| 755 |
+
"step": 95
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.92,
|
| 759 |
+
"grad_norm": 7.311519145965576,
|
| 760 |
+
"learning_rate": 2.7368421052631583e-05,
|
| 761 |
+
"loss": 0.0235,
|
| 762 |
+
"step": 96
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.94,
|
| 766 |
+
"grad_norm": 53.3145866394043,
|
| 767 |
+
"learning_rate": 2.710526315789474e-05,
|
| 768 |
+
"loss": 0.1186,
|
| 769 |
+
"step": 97
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.96,
|
| 773 |
+
"grad_norm": 13.420350074768066,
|
| 774 |
+
"learning_rate": 2.6842105263157896e-05,
|
| 775 |
+
"loss": 0.2332,
|
| 776 |
+
"step": 98
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.98,
|
| 780 |
+
"grad_norm": 13.3897123336792,
|
| 781 |
+
"learning_rate": 2.6578947368421052e-05,
|
| 782 |
+
"loss": 0.0759,
|
| 783 |
+
"step": 99
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.0,
|
| 787 |
+
"grad_norm": 1.8416231870651245,
|
| 788 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 789 |
+
"loss": 0.0076,
|
| 790 |
+
"step": 100
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.0,
|
| 794 |
+
"eval_accuracy": 0.993421052631579,
|
| 795 |
+
"eval_loss": 0.027568014338612556,
|
| 796 |
+
"eval_runtime": 7.0992,
|
| 797 |
+
"eval_samples_per_second": 42.822,
|
| 798 |
+
"eval_steps_per_second": 1.127,
|
| 799 |
+
"step": 100
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.02,
|
| 803 |
+
"grad_norm": 6.973387241363525,
|
| 804 |
+
"learning_rate": 2.605263157894737e-05,
|
| 805 |
+
"loss": 0.0255,
|
| 806 |
+
"step": 101
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.04,
|
| 810 |
+
"grad_norm": 0.33660009503364563,
|
| 811 |
+
"learning_rate": 2.578947368421053e-05,
|
| 812 |
+
"loss": 0.0012,
|
| 813 |
+
"step": 102
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.06,
|
| 817 |
+
"grad_norm": 11.269298553466797,
|
| 818 |
+
"learning_rate": 2.5526315789473688e-05,
|
| 819 |
+
"loss": 0.1097,
|
| 820 |
+
"step": 103
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.08,
|
| 824 |
+
"grad_norm": 7.602676868438721,
|
| 825 |
+
"learning_rate": 2.5263157894736845e-05,
|
| 826 |
+
"loss": 0.034,
|
| 827 |
+
"step": 104
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 2.1,
|
| 831 |
+
"grad_norm": 7.900285720825195,
|
| 832 |
+
"learning_rate": 2.5e-05,
|
| 833 |
+
"loss": 0.0399,
|
| 834 |
+
"step": 105
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.12,
|
| 838 |
+
"grad_norm": 4.07220983505249,
|
| 839 |
+
"learning_rate": 2.4736842105263158e-05,
|
| 840 |
+
"loss": 0.0055,
|
| 841 |
+
"step": 106
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.14,
|
| 845 |
+
"grad_norm": 0.09085848927497864,
|
| 846 |
+
"learning_rate": 2.4473684210526318e-05,
|
| 847 |
+
"loss": 0.0005,
|
| 848 |
+
"step": 107
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.16,
|
| 852 |
+
"grad_norm": 0.5666660666465759,
|
| 853 |
+
"learning_rate": 2.4210526315789474e-05,
|
| 854 |
+
"loss": 0.0029,
|
| 855 |
+
"step": 108
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 2.18,
|
| 859 |
+
"grad_norm": 12.90062427520752,
|
| 860 |
+
"learning_rate": 2.394736842105263e-05,
|
| 861 |
+
"loss": 0.051,
|
| 862 |
+
"step": 109
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 2.2,
|
| 866 |
+
"grad_norm": 0.24952377378940582,
|
| 867 |
+
"learning_rate": 2.368421052631579e-05,
|
| 868 |
+
"loss": 0.0007,
|
| 869 |
+
"step": 110
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 2.2,
|
| 873 |
+
"eval_accuracy": 0.9835526315789473,
|
| 874 |
+
"eval_loss": 0.05723079293966293,
|
| 875 |
+
"eval_runtime": 7.2757,
|
| 876 |
+
"eval_samples_per_second": 41.783,
|
| 877 |
+
"eval_steps_per_second": 1.1,
|
| 878 |
+
"step": 110
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.22,
|
| 882 |
+
"grad_norm": 2.654731512069702,
|
| 883 |
+
"learning_rate": 2.342105263157895e-05,
|
| 884 |
+
"loss": 0.0036,
|
| 885 |
+
"step": 111
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.24,
|
| 889 |
+
"grad_norm": 0.5114130973815918,
|
| 890 |
+
"learning_rate": 2.3157894736842107e-05,
|
| 891 |
+
"loss": 0.0012,
|
| 892 |
+
"step": 112
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 2.26,
|
| 896 |
+
"grad_norm": 9.55749797821045,
|
| 897 |
+
"learning_rate": 2.2894736842105263e-05,
|
| 898 |
+
"loss": 0.0195,
|
| 899 |
+
"step": 113
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.28,
|
| 903 |
+
"grad_norm": 11.102432250976562,
|
| 904 |
+
"learning_rate": 2.2631578947368423e-05,
|
| 905 |
+
"loss": 0.0204,
|
| 906 |
+
"step": 114
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 2.3,
|
| 910 |
+
"grad_norm": 2.684544324874878,
|
| 911 |
+
"learning_rate": 2.236842105263158e-05,
|
| 912 |
+
"loss": 0.0047,
|
| 913 |
+
"step": 115
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.32,
|
| 917 |
+
"grad_norm": 0.22236469388008118,
|
| 918 |
+
"learning_rate": 2.2105263157894736e-05,
|
| 919 |
+
"loss": 0.0008,
|
| 920 |
+
"step": 116
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 2.34,
|
| 924 |
+
"grad_norm": 8.926434516906738,
|
| 925 |
+
"learning_rate": 2.1842105263157896e-05,
|
| 926 |
+
"loss": 0.0223,
|
| 927 |
+
"step": 117
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.36,
|
| 931 |
+
"grad_norm": 3.3315141201019287,
|
| 932 |
+
"learning_rate": 2.1578947368421053e-05,
|
| 933 |
+
"loss": 0.0117,
|
| 934 |
+
"step": 118
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 2.38,
|
| 938 |
+
"grad_norm": 10.430353164672852,
|
| 939 |
+
"learning_rate": 2.1315789473684212e-05,
|
| 940 |
+
"loss": 0.0303,
|
| 941 |
+
"step": 119
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 2.4,
|
| 945 |
+
"grad_norm": 0.15297500789165497,
|
| 946 |
+
"learning_rate": 2.105263157894737e-05,
|
| 947 |
+
"loss": 0.0008,
|
| 948 |
+
"step": 120
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 2.4,
|
| 952 |
+
"eval_accuracy": 0.9967105263157895,
|
| 953 |
+
"eval_loss": 0.009093067608773708,
|
| 954 |
+
"eval_runtime": 6.8131,
|
| 955 |
+
"eval_samples_per_second": 44.62,
|
| 956 |
+
"eval_steps_per_second": 1.174,
|
| 957 |
+
"step": 120
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.42,
|
| 961 |
+
"grad_norm": 0.6949332356452942,
|
| 962 |
+
"learning_rate": 2.078947368421053e-05,
|
| 963 |
+
"loss": 0.0018,
|
| 964 |
+
"step": 121
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 2.44,
|
| 968 |
+
"grad_norm": 0.10584941506385803,
|
| 969 |
+
"learning_rate": 2.0526315789473685e-05,
|
| 970 |
+
"loss": 0.0006,
|
| 971 |
+
"step": 122
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.46,
|
| 975 |
+
"grad_norm": 2.782750368118286,
|
| 976 |
+
"learning_rate": 2.0263157894736842e-05,
|
| 977 |
+
"loss": 0.0064,
|
| 978 |
+
"step": 123
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 2.48,
|
| 982 |
+
"grad_norm": 0.1843283325433731,
|
| 983 |
+
"learning_rate": 2e-05,
|
| 984 |
+
"loss": 0.0007,
|
| 985 |
+
"step": 124
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 2.5,
|
| 989 |
+
"grad_norm": 19.077423095703125,
|
| 990 |
+
"learning_rate": 1.9736842105263158e-05,
|
| 991 |
+
"loss": 0.1095,
|
| 992 |
+
"step": 125
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.52,
|
| 996 |
+
"grad_norm": 9.631857872009277,
|
| 997 |
+
"learning_rate": 1.9473684210526315e-05,
|
| 998 |
+
"loss": 0.0286,
|
| 999 |
+
"step": 126
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.54,
|
| 1003 |
+
"grad_norm": 2.46711802482605,
|
| 1004 |
+
"learning_rate": 1.9210526315789474e-05,
|
| 1005 |
+
"loss": 0.0042,
|
| 1006 |
+
"step": 127
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 2.56,
|
| 1010 |
+
"grad_norm": 0.9471967220306396,
|
| 1011 |
+
"learning_rate": 1.8947368421052634e-05,
|
| 1012 |
+
"loss": 0.0038,
|
| 1013 |
+
"step": 128
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 2.58,
|
| 1017 |
+
"grad_norm": 0.1509537547826767,
|
| 1018 |
+
"learning_rate": 1.868421052631579e-05,
|
| 1019 |
+
"loss": 0.0009,
|
| 1020 |
+
"step": 129
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 2.6,
|
| 1024 |
+
"grad_norm": 16.95519256591797,
|
| 1025 |
+
"learning_rate": 1.8421052631578947e-05,
|
| 1026 |
+
"loss": 0.1277,
|
| 1027 |
+
"step": 130
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 2.6,
|
| 1031 |
+
"eval_accuracy": 0.9967105263157895,
|
| 1032 |
+
"eval_loss": 0.004848334938287735,
|
| 1033 |
+
"eval_runtime": 7.1393,
|
| 1034 |
+
"eval_samples_per_second": 42.581,
|
| 1035 |
+
"eval_steps_per_second": 1.121,
|
| 1036 |
+
"step": 130
|
| 1037 |
+
}
|
| 1038 |
+
],
|
| 1039 |
+
"logging_steps": 1,
|
| 1040 |
+
"max_steps": 200,
|
| 1041 |
+
"num_input_tokens_seen": 0,
|
| 1042 |
+
"num_train_epochs": 4,
|
| 1043 |
+
"save_steps": 10,
|
| 1044 |
+
"total_flos": 4.0410637592938906e+17,
|
| 1045 |
+
"train_batch_size": 40,
|
| 1046 |
+
"trial_name": null,
|
| 1047 |
+
"trial_params": null
|
| 1048 |
+
}
|
checkpoint-130/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d12326b817ccf86422554696b6e15d756b4174e67e313ecfbe14aadf3bb73244
|
| 3 |
+
size 4856
|
checkpoint-190/config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Organika/sdxl-detector",
|
| 3 |
+
"_num_labels": 2,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"SwinForImageClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.0,
|
| 8 |
+
"depths": [
|
| 9 |
+
2,
|
| 10 |
+
2,
|
| 11 |
+
18,
|
| 12 |
+
2
|
| 13 |
+
],
|
| 14 |
+
"drop_path_rate": 0.1,
|
| 15 |
+
"embed_dim": 128,
|
| 16 |
+
"encoder_stride": 32,
|
| 17 |
+
"hidden_act": "gelu",
|
| 18 |
+
"hidden_dropout_prob": 0.0,
|
| 19 |
+
"hidden_size": 1024,
|
| 20 |
+
"id2label": {
|
| 21 |
+
"0": "artificial",
|
| 22 |
+
"1": "human"
|
| 23 |
+
},
|
| 24 |
+
"image_size": 224,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"label2id": {
|
| 27 |
+
"artificial": 0,
|
| 28 |
+
"human": 1
|
| 29 |
+
},
|
| 30 |
+
"layer_norm_eps": 1e-05,
|
| 31 |
+
"max_length": 128,
|
| 32 |
+
"mlp_ratio": 4.0,
|
| 33 |
+
"model_type": "swin",
|
| 34 |
+
"num_channels": 3,
|
| 35 |
+
"num_heads": [
|
| 36 |
+
4,
|
| 37 |
+
8,
|
| 38 |
+
16,
|
| 39 |
+
32
|
| 40 |
+
],
|
| 41 |
+
"num_layers": 4,
|
| 42 |
+
"out_features": [
|
| 43 |
+
"stage4"
|
| 44 |
+
],
|
| 45 |
+
"out_indices": [
|
| 46 |
+
4
|
| 47 |
+
],
|
| 48 |
+
"padding": "max_length",
|
| 49 |
+
"patch_size": 4,
|
| 50 |
+
"path_norm": true,
|
| 51 |
+
"problem_type": "single_label_classification",
|
| 52 |
+
"qkv_bias": true,
|
| 53 |
+
"stage_names": [
|
| 54 |
+
"stem",
|
| 55 |
+
"stage1",
|
| 56 |
+
"stage2",
|
| 57 |
+
"stage3",
|
| 58 |
+
"stage4"
|
| 59 |
+
],
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.39.3",
|
| 62 |
+
"use_absolute_embeddings": false,
|
| 63 |
+
"window_size": 7
|
| 64 |
+
}
|
checkpoint-190/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a362f01847b6f823cfdf2f60859ea678b90242148dc1eb2218a5bf6ee5fa54bd
|
| 3 |
+
size 347498816
|
checkpoint-190/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a961a977a85ba1ef4943ab2cc748ffd5de8ef9677eb7c5f0b3c752dd55f1986
|
| 3 |
+
size 694317645
|
checkpoint-190/preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.485,
|
| 21 |
+
0.456,
|
| 22 |
+
0.406
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.229,
|
| 27 |
+
0.224,
|
| 28 |
+
0.225
|
| 29 |
+
],
|
| 30 |
+
"resample": 3,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
checkpoint-190/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2b872ea464d68185e024295eb3a72fc1615395a5617c66d16dbb8cbf3226ee7
|
| 3 |
+
size 14244
|
checkpoint-190/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:359e75abc7a98a86bc32539a93eabc5596c39ca4b881df61b9957f610c6eecb0
|
| 3 |
+
size 1064
|
checkpoint-190/trainer_state.json
ADDED
|
@@ -0,0 +1,1522 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.004848334938287735,
|
| 3 |
+
"best_model_checkpoint": "ai_vs_real_image_detection/checkpoint-130",
|
| 4 |
+
"epoch": 3.8,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 190,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 52.89345169067383,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 2.2763,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04,
|
| 20 |
+
"grad_norm": 62.66461944580078,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.8254,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06,
|
| 27 |
+
"grad_norm": 66.1015625,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 1.8906,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08,
|
| 34 |
+
"grad_norm": 56.700233459472656,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 1.3529,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1,
|
| 41 |
+
"grad_norm": 66.10917663574219,
|
| 42 |
+
"learning_rate": 2.5e-05,
|
| 43 |
+
"loss": 1.4672,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12,
|
| 48 |
+
"grad_norm": 52.68284225463867,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 1.0578,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14,
|
| 55 |
+
"grad_norm": 42.4580078125,
|
| 56 |
+
"learning_rate": 3.5e-05,
|
| 57 |
+
"loss": 1.1221,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.16,
|
| 62 |
+
"grad_norm": 64.87982940673828,
|
| 63 |
+
"learning_rate": 4e-05,
|
| 64 |
+
"loss": 0.9313,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.18,
|
| 69 |
+
"grad_norm": 56.227081298828125,
|
| 70 |
+
"learning_rate": 4.5e-05,
|
| 71 |
+
"loss": 0.644,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2,
|
| 76 |
+
"grad_norm": 13.719300270080566,
|
| 77 |
+
"learning_rate": 5e-05,
|
| 78 |
+
"loss": 0.4305,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2,
|
| 83 |
+
"eval_accuracy": 0.9144736842105263,
|
| 84 |
+
"eval_loss": 0.19980163872241974,
|
| 85 |
+
"eval_runtime": 6.726,
|
| 86 |
+
"eval_samples_per_second": 45.198,
|
| 87 |
+
"eval_steps_per_second": 1.189,
|
| 88 |
+
"step": 10
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.22,
|
| 92 |
+
"grad_norm": 16.044822692871094,
|
| 93 |
+
"learning_rate": 4.973684210526316e-05,
|
| 94 |
+
"loss": 0.3892,
|
| 95 |
+
"step": 11
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.24,
|
| 99 |
+
"grad_norm": 15.53343391418457,
|
| 100 |
+
"learning_rate": 4.9473684210526315e-05,
|
| 101 |
+
"loss": 0.3683,
|
| 102 |
+
"step": 12
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.26,
|
| 106 |
+
"grad_norm": 13.435264587402344,
|
| 107 |
+
"learning_rate": 4.921052631578947e-05,
|
| 108 |
+
"loss": 0.3594,
|
| 109 |
+
"step": 13
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.28,
|
| 113 |
+
"grad_norm": 17.45833969116211,
|
| 114 |
+
"learning_rate": 4.8947368421052635e-05,
|
| 115 |
+
"loss": 0.5304,
|
| 116 |
+
"step": 14
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.3,
|
| 120 |
+
"grad_norm": 14.26673412322998,
|
| 121 |
+
"learning_rate": 4.868421052631579e-05,
|
| 122 |
+
"loss": 0.4804,
|
| 123 |
+
"step": 15
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.32,
|
| 127 |
+
"grad_norm": 9.26322078704834,
|
| 128 |
+
"learning_rate": 4.842105263157895e-05,
|
| 129 |
+
"loss": 0.3633,
|
| 130 |
+
"step": 16
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.34,
|
| 134 |
+
"grad_norm": 8.355168342590332,
|
| 135 |
+
"learning_rate": 4.8157894736842105e-05,
|
| 136 |
+
"loss": 0.4711,
|
| 137 |
+
"step": 17
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.36,
|
| 141 |
+
"grad_norm": 8.733236312866211,
|
| 142 |
+
"learning_rate": 4.789473684210526e-05,
|
| 143 |
+
"loss": 0.3954,
|
| 144 |
+
"step": 18
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.38,
|
| 148 |
+
"grad_norm": 5.385647296905518,
|
| 149 |
+
"learning_rate": 4.7631578947368424e-05,
|
| 150 |
+
"loss": 0.1943,
|
| 151 |
+
"step": 19
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.4,
|
| 155 |
+
"grad_norm": 9.892962455749512,
|
| 156 |
+
"learning_rate": 4.736842105263158e-05,
|
| 157 |
+
"loss": 0.3209,
|
| 158 |
+
"step": 20
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.4,
|
| 162 |
+
"eval_accuracy": 0.8881578947368421,
|
| 163 |
+
"eval_loss": 0.24087880551815033,
|
| 164 |
+
"eval_runtime": 6.8213,
|
| 165 |
+
"eval_samples_per_second": 44.566,
|
| 166 |
+
"eval_steps_per_second": 1.173,
|
| 167 |
+
"step": 20
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.42,
|
| 171 |
+
"grad_norm": 5.46142578125,
|
| 172 |
+
"learning_rate": 4.7105263157894744e-05,
|
| 173 |
+
"loss": 0.1505,
|
| 174 |
+
"step": 21
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.44,
|
| 178 |
+
"grad_norm": 9.863460540771484,
|
| 179 |
+
"learning_rate": 4.68421052631579e-05,
|
| 180 |
+
"loss": 0.2481,
|
| 181 |
+
"step": 22
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.46,
|
| 185 |
+
"grad_norm": 7.595032691955566,
|
| 186 |
+
"learning_rate": 4.657894736842106e-05,
|
| 187 |
+
"loss": 0.2001,
|
| 188 |
+
"step": 23
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.48,
|
| 192 |
+
"grad_norm": 17.375865936279297,
|
| 193 |
+
"learning_rate": 4.6315789473684214e-05,
|
| 194 |
+
"loss": 0.336,
|
| 195 |
+
"step": 24
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.5,
|
| 199 |
+
"grad_norm": 5.597734451293945,
|
| 200 |
+
"learning_rate": 4.605263157894737e-05,
|
| 201 |
+
"loss": 0.1395,
|
| 202 |
+
"step": 25
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.52,
|
| 206 |
+
"grad_norm": 7.56685209274292,
|
| 207 |
+
"learning_rate": 4.5789473684210527e-05,
|
| 208 |
+
"loss": 0.1739,
|
| 209 |
+
"step": 26
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.54,
|
| 213 |
+
"grad_norm": 13.769929885864258,
|
| 214 |
+
"learning_rate": 4.552631578947369e-05,
|
| 215 |
+
"loss": 0.1959,
|
| 216 |
+
"step": 27
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.56,
|
| 220 |
+
"grad_norm": 10.677892684936523,
|
| 221 |
+
"learning_rate": 4.5263157894736846e-05,
|
| 222 |
+
"loss": 0.2015,
|
| 223 |
+
"step": 28
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.58,
|
| 227 |
+
"grad_norm": 3.645374059677124,
|
| 228 |
+
"learning_rate": 4.5e-05,
|
| 229 |
+
"loss": 0.0576,
|
| 230 |
+
"step": 29
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.6,
|
| 234 |
+
"grad_norm": 12.538244247436523,
|
| 235 |
+
"learning_rate": 4.473684210526316e-05,
|
| 236 |
+
"loss": 0.324,
|
| 237 |
+
"step": 30
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.6,
|
| 241 |
+
"eval_accuracy": 0.9802631578947368,
|
| 242 |
+
"eval_loss": 0.056003469973802567,
|
| 243 |
+
"eval_runtime": 6.4198,
|
| 244 |
+
"eval_samples_per_second": 47.353,
|
| 245 |
+
"eval_steps_per_second": 1.246,
|
| 246 |
+
"step": 30
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.62,
|
| 250 |
+
"grad_norm": 8.700516700744629,
|
| 251 |
+
"learning_rate": 4.4473684210526316e-05,
|
| 252 |
+
"loss": 0.1199,
|
| 253 |
+
"step": 31
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.64,
|
| 257 |
+
"grad_norm": 6.926602840423584,
|
| 258 |
+
"learning_rate": 4.421052631578947e-05,
|
| 259 |
+
"loss": 0.1471,
|
| 260 |
+
"step": 32
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.66,
|
| 264 |
+
"grad_norm": 8.461554527282715,
|
| 265 |
+
"learning_rate": 4.394736842105263e-05,
|
| 266 |
+
"loss": 0.0649,
|
| 267 |
+
"step": 33
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.68,
|
| 271 |
+
"grad_norm": 7.129871845245361,
|
| 272 |
+
"learning_rate": 4.368421052631579e-05,
|
| 273 |
+
"loss": 0.1167,
|
| 274 |
+
"step": 34
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.7,
|
| 278 |
+
"grad_norm": 20.945388793945312,
|
| 279 |
+
"learning_rate": 4.342105263157895e-05,
|
| 280 |
+
"loss": 0.3279,
|
| 281 |
+
"step": 35
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.72,
|
| 285 |
+
"grad_norm": 4.466580390930176,
|
| 286 |
+
"learning_rate": 4.3157894736842105e-05,
|
| 287 |
+
"loss": 0.0517,
|
| 288 |
+
"step": 36
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.74,
|
| 292 |
+
"grad_norm": 9.97375202178955,
|
| 293 |
+
"learning_rate": 4.289473684210527e-05,
|
| 294 |
+
"loss": 0.1241,
|
| 295 |
+
"step": 37
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.76,
|
| 299 |
+
"grad_norm": 6.459863662719727,
|
| 300 |
+
"learning_rate": 4.2631578947368425e-05,
|
| 301 |
+
"loss": 0.0859,
|
| 302 |
+
"step": 38
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.78,
|
| 306 |
+
"grad_norm": 11.704880714416504,
|
| 307 |
+
"learning_rate": 4.236842105263158e-05,
|
| 308 |
+
"loss": 0.1252,
|
| 309 |
+
"step": 39
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.8,
|
| 313 |
+
"grad_norm": 7.90261697769165,
|
| 314 |
+
"learning_rate": 4.210526315789474e-05,
|
| 315 |
+
"loss": 0.0541,
|
| 316 |
+
"step": 40
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.8,
|
| 320 |
+
"eval_accuracy": 0.9671052631578947,
|
| 321 |
+
"eval_loss": 0.06965293735265732,
|
| 322 |
+
"eval_runtime": 6.6281,
|
| 323 |
+
"eval_samples_per_second": 45.865,
|
| 324 |
+
"eval_steps_per_second": 1.207,
|
| 325 |
+
"step": 40
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.82,
|
| 329 |
+
"grad_norm": 2.8429932594299316,
|
| 330 |
+
"learning_rate": 4.18421052631579e-05,
|
| 331 |
+
"loss": 0.0325,
|
| 332 |
+
"step": 41
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.84,
|
| 336 |
+
"grad_norm": 10.533605575561523,
|
| 337 |
+
"learning_rate": 4.157894736842106e-05,
|
| 338 |
+
"loss": 0.1836,
|
| 339 |
+
"step": 42
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.86,
|
| 343 |
+
"grad_norm": 12.08105754852295,
|
| 344 |
+
"learning_rate": 4.1315789473684214e-05,
|
| 345 |
+
"loss": 0.224,
|
| 346 |
+
"step": 43
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.88,
|
| 350 |
+
"grad_norm": 9.006488800048828,
|
| 351 |
+
"learning_rate": 4.105263157894737e-05,
|
| 352 |
+
"loss": 0.2139,
|
| 353 |
+
"step": 44
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9,
|
| 357 |
+
"grad_norm": 7.3875956535339355,
|
| 358 |
+
"learning_rate": 4.078947368421053e-05,
|
| 359 |
+
"loss": 0.088,
|
| 360 |
+
"step": 45
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.92,
|
| 364 |
+
"grad_norm": 13.019746780395508,
|
| 365 |
+
"learning_rate": 4.0526315789473684e-05,
|
| 366 |
+
"loss": 0.2363,
|
| 367 |
+
"step": 46
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.94,
|
| 371 |
+
"grad_norm": 37.95082473754883,
|
| 372 |
+
"learning_rate": 4.026315789473684e-05,
|
| 373 |
+
"loss": 0.3728,
|
| 374 |
+
"step": 47
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.96,
|
| 378 |
+
"grad_norm": 13.99156665802002,
|
| 379 |
+
"learning_rate": 4e-05,
|
| 380 |
+
"loss": 0.1376,
|
| 381 |
+
"step": 48
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.98,
|
| 385 |
+
"grad_norm": 5.063250541687012,
|
| 386 |
+
"learning_rate": 3.973684210526316e-05,
|
| 387 |
+
"loss": 0.0832,
|
| 388 |
+
"step": 49
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"grad_norm": 13.046541213989258,
|
| 393 |
+
"learning_rate": 3.9473684210526316e-05,
|
| 394 |
+
"loss": 0.1096,
|
| 395 |
+
"step": 50
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.0,
|
| 399 |
+
"eval_accuracy": 0.9572368421052632,
|
| 400 |
+
"eval_loss": 0.12372467666864395,
|
| 401 |
+
"eval_runtime": 6.8765,
|
| 402 |
+
"eval_samples_per_second": 44.209,
|
| 403 |
+
"eval_steps_per_second": 1.163,
|
| 404 |
+
"step": 50
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.02,
|
| 408 |
+
"grad_norm": 4.865447998046875,
|
| 409 |
+
"learning_rate": 3.921052631578947e-05,
|
| 410 |
+
"loss": 0.0366,
|
| 411 |
+
"step": 51
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.04,
|
| 415 |
+
"grad_norm": 12.430005073547363,
|
| 416 |
+
"learning_rate": 3.894736842105263e-05,
|
| 417 |
+
"loss": 0.1301,
|
| 418 |
+
"step": 52
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 1.06,
|
| 422 |
+
"grad_norm": 2.1291539669036865,
|
| 423 |
+
"learning_rate": 3.868421052631579e-05,
|
| 424 |
+
"loss": 0.0168,
|
| 425 |
+
"step": 53
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.08,
|
| 429 |
+
"grad_norm": 1.6922448873519897,
|
| 430 |
+
"learning_rate": 3.842105263157895e-05,
|
| 431 |
+
"loss": 0.0074,
|
| 432 |
+
"step": 54
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.1,
|
| 436 |
+
"grad_norm": 5.851077556610107,
|
| 437 |
+
"learning_rate": 3.815789473684211e-05,
|
| 438 |
+
"loss": 0.0216,
|
| 439 |
+
"step": 55
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.12,
|
| 443 |
+
"grad_norm": 21.35593605041504,
|
| 444 |
+
"learning_rate": 3.789473684210527e-05,
|
| 445 |
+
"loss": 0.1171,
|
| 446 |
+
"step": 56
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.14,
|
| 450 |
+
"grad_norm": 9.791412353515625,
|
| 451 |
+
"learning_rate": 3.7631578947368425e-05,
|
| 452 |
+
"loss": 0.1688,
|
| 453 |
+
"step": 57
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.16,
|
| 457 |
+
"grad_norm": 3.2542812824249268,
|
| 458 |
+
"learning_rate": 3.736842105263158e-05,
|
| 459 |
+
"loss": 0.0236,
|
| 460 |
+
"step": 58
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.18,
|
| 464 |
+
"grad_norm": 9.557928085327148,
|
| 465 |
+
"learning_rate": 3.710526315789474e-05,
|
| 466 |
+
"loss": 0.1782,
|
| 467 |
+
"step": 59
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.2,
|
| 471 |
+
"grad_norm": 12.514122009277344,
|
| 472 |
+
"learning_rate": 3.6842105263157895e-05,
|
| 473 |
+
"loss": 0.0905,
|
| 474 |
+
"step": 60
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.2,
|
| 478 |
+
"eval_accuracy": 0.9967105263157895,
|
| 479 |
+
"eval_loss": 0.027645185589790344,
|
| 480 |
+
"eval_runtime": 7.1325,
|
| 481 |
+
"eval_samples_per_second": 42.622,
|
| 482 |
+
"eval_steps_per_second": 1.122,
|
| 483 |
+
"step": 60
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.22,
|
| 487 |
+
"grad_norm": 4.434255599975586,
|
| 488 |
+
"learning_rate": 3.657894736842106e-05,
|
| 489 |
+
"loss": 0.0276,
|
| 490 |
+
"step": 61
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.24,
|
| 494 |
+
"grad_norm": 6.304673671722412,
|
| 495 |
+
"learning_rate": 3.6315789473684214e-05,
|
| 496 |
+
"loss": 0.0508,
|
| 497 |
+
"step": 62
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.26,
|
| 501 |
+
"grad_norm": 3.844200372695923,
|
| 502 |
+
"learning_rate": 3.605263157894737e-05,
|
| 503 |
+
"loss": 0.0199,
|
| 504 |
+
"step": 63
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.28,
|
| 508 |
+
"grad_norm": 0.8185235261917114,
|
| 509 |
+
"learning_rate": 3.578947368421053e-05,
|
| 510 |
+
"loss": 0.0069,
|
| 511 |
+
"step": 64
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.3,
|
| 515 |
+
"grad_norm": 6.909903049468994,
|
| 516 |
+
"learning_rate": 3.5526315789473684e-05,
|
| 517 |
+
"loss": 0.0597,
|
| 518 |
+
"step": 65
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.32,
|
| 522 |
+
"grad_norm": 3.5652379989624023,
|
| 523 |
+
"learning_rate": 3.526315789473684e-05,
|
| 524 |
+
"loss": 0.0141,
|
| 525 |
+
"step": 66
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.34,
|
| 529 |
+
"grad_norm": 11.72857666015625,
|
| 530 |
+
"learning_rate": 3.5e-05,
|
| 531 |
+
"loss": 0.1865,
|
| 532 |
+
"step": 67
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.36,
|
| 536 |
+
"grad_norm": 10.541884422302246,
|
| 537 |
+
"learning_rate": 3.473684210526316e-05,
|
| 538 |
+
"loss": 0.0606,
|
| 539 |
+
"step": 68
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.38,
|
| 543 |
+
"grad_norm": 5.210526943206787,
|
| 544 |
+
"learning_rate": 3.447368421052632e-05,
|
| 545 |
+
"loss": 0.0276,
|
| 546 |
+
"step": 69
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.4,
|
| 550 |
+
"grad_norm": 1.0411546230316162,
|
| 551 |
+
"learning_rate": 3.421052631578947e-05,
|
| 552 |
+
"loss": 0.009,
|
| 553 |
+
"step": 70
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.4,
|
| 557 |
+
"eval_accuracy": 0.9901315789473685,
|
| 558 |
+
"eval_loss": 0.020390018820762634,
|
| 559 |
+
"eval_runtime": 7.2971,
|
| 560 |
+
"eval_samples_per_second": 41.66,
|
| 561 |
+
"eval_steps_per_second": 1.096,
|
| 562 |
+
"step": 70
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.42,
|
| 566 |
+
"grad_norm": 0.7153753638267517,
|
| 567 |
+
"learning_rate": 3.3947368421052636e-05,
|
| 568 |
+
"loss": 0.0049,
|
| 569 |
+
"step": 71
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.44,
|
| 573 |
+
"grad_norm": 4.828935623168945,
|
| 574 |
+
"learning_rate": 3.368421052631579e-05,
|
| 575 |
+
"loss": 0.0297,
|
| 576 |
+
"step": 72
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.46,
|
| 580 |
+
"grad_norm": 9.103313446044922,
|
| 581 |
+
"learning_rate": 3.342105263157895e-05,
|
| 582 |
+
"loss": 0.071,
|
| 583 |
+
"step": 73
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.48,
|
| 587 |
+
"grad_norm": 7.635372161865234,
|
| 588 |
+
"learning_rate": 3.3157894736842106e-05,
|
| 589 |
+
"loss": 0.0363,
|
| 590 |
+
"step": 74
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.5,
|
| 594 |
+
"grad_norm": 8.57288646697998,
|
| 595 |
+
"learning_rate": 3.289473684210527e-05,
|
| 596 |
+
"loss": 0.0566,
|
| 597 |
+
"step": 75
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.52,
|
| 601 |
+
"grad_norm": 8.89769172668457,
|
| 602 |
+
"learning_rate": 3.2631578947368426e-05,
|
| 603 |
+
"loss": 0.0213,
|
| 604 |
+
"step": 76
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.54,
|
| 608 |
+
"grad_norm": 2.804738998413086,
|
| 609 |
+
"learning_rate": 3.236842105263158e-05,
|
| 610 |
+
"loss": 0.0138,
|
| 611 |
+
"step": 77
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.56,
|
| 615 |
+
"grad_norm": 4.841202259063721,
|
| 616 |
+
"learning_rate": 3.210526315789474e-05,
|
| 617 |
+
"loss": 0.0181,
|
| 618 |
+
"step": 78
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.58,
|
| 622 |
+
"grad_norm": 7.109033584594727,
|
| 623 |
+
"learning_rate": 3.1842105263157895e-05,
|
| 624 |
+
"loss": 0.0332,
|
| 625 |
+
"step": 79
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.6,
|
| 629 |
+
"grad_norm": 13.540718078613281,
|
| 630 |
+
"learning_rate": 3.157894736842105e-05,
|
| 631 |
+
"loss": 0.1667,
|
| 632 |
+
"step": 80
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.6,
|
| 636 |
+
"eval_accuracy": 0.9703947368421053,
|
| 637 |
+
"eval_loss": 0.10031529515981674,
|
| 638 |
+
"eval_runtime": 6.7491,
|
| 639 |
+
"eval_samples_per_second": 45.043,
|
| 640 |
+
"eval_steps_per_second": 1.185,
|
| 641 |
+
"step": 80
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.62,
|
| 645 |
+
"grad_norm": 30.279451370239258,
|
| 646 |
+
"learning_rate": 3.131578947368421e-05,
|
| 647 |
+
"loss": 0.1139,
|
| 648 |
+
"step": 81
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.64,
|
| 652 |
+
"grad_norm": 7.229616165161133,
|
| 653 |
+
"learning_rate": 3.105263157894737e-05,
|
| 654 |
+
"loss": 0.0284,
|
| 655 |
+
"step": 82
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.66,
|
| 659 |
+
"grad_norm": 0.327736496925354,
|
| 660 |
+
"learning_rate": 3.078947368421053e-05,
|
| 661 |
+
"loss": 0.0032,
|
| 662 |
+
"step": 83
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.68,
|
| 666 |
+
"grad_norm": 0.15309709310531616,
|
| 667 |
+
"learning_rate": 3.0526315789473684e-05,
|
| 668 |
+
"loss": 0.0011,
|
| 669 |
+
"step": 84
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.7,
|
| 673 |
+
"grad_norm": 13.175533294677734,
|
| 674 |
+
"learning_rate": 3.0263157894736844e-05,
|
| 675 |
+
"loss": 0.039,
|
| 676 |
+
"step": 85
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.72,
|
| 680 |
+
"grad_norm": 24.578662872314453,
|
| 681 |
+
"learning_rate": 3e-05,
|
| 682 |
+
"loss": 0.1834,
|
| 683 |
+
"step": 86
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.74,
|
| 687 |
+
"grad_norm": 6.690978527069092,
|
| 688 |
+
"learning_rate": 2.9736842105263157e-05,
|
| 689 |
+
"loss": 0.0214,
|
| 690 |
+
"step": 87
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.76,
|
| 694 |
+
"grad_norm": 21.097963333129883,
|
| 695 |
+
"learning_rate": 2.9473684210526314e-05,
|
| 696 |
+
"loss": 0.1569,
|
| 697 |
+
"step": 88
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.78,
|
| 701 |
+
"grad_norm": 7.984706401824951,
|
| 702 |
+
"learning_rate": 2.9210526315789477e-05,
|
| 703 |
+
"loss": 0.0326,
|
| 704 |
+
"step": 89
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.8,
|
| 708 |
+
"grad_norm": 11.008134841918945,
|
| 709 |
+
"learning_rate": 2.8947368421052634e-05,
|
| 710 |
+
"loss": 0.0602,
|
| 711 |
+
"step": 90
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.8,
|
| 715 |
+
"eval_accuracy": 0.9868421052631579,
|
| 716 |
+
"eval_loss": 0.052807241678237915,
|
| 717 |
+
"eval_runtime": 7.1111,
|
| 718 |
+
"eval_samples_per_second": 42.75,
|
| 719 |
+
"eval_steps_per_second": 1.125,
|
| 720 |
+
"step": 90
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.82,
|
| 724 |
+
"grad_norm": 21.149032592773438,
|
| 725 |
+
"learning_rate": 2.868421052631579e-05,
|
| 726 |
+
"loss": 0.0265,
|
| 727 |
+
"step": 91
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.84,
|
| 731 |
+
"grad_norm": 3.4383814334869385,
|
| 732 |
+
"learning_rate": 2.842105263157895e-05,
|
| 733 |
+
"loss": 0.0059,
|
| 734 |
+
"step": 92
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.86,
|
| 738 |
+
"grad_norm": 6.291805267333984,
|
| 739 |
+
"learning_rate": 2.8157894736842106e-05,
|
| 740 |
+
"loss": 0.0177,
|
| 741 |
+
"step": 93
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 1.88,
|
| 745 |
+
"grad_norm": 1.5197041034698486,
|
| 746 |
+
"learning_rate": 2.7894736842105263e-05,
|
| 747 |
+
"loss": 0.0065,
|
| 748 |
+
"step": 94
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 1.9,
|
| 752 |
+
"grad_norm": 0.47119140625,
|
| 753 |
+
"learning_rate": 2.7631578947368426e-05,
|
| 754 |
+
"loss": 0.0018,
|
| 755 |
+
"step": 95
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.92,
|
| 759 |
+
"grad_norm": 7.311519145965576,
|
| 760 |
+
"learning_rate": 2.7368421052631583e-05,
|
| 761 |
+
"loss": 0.0235,
|
| 762 |
+
"step": 96
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.94,
|
| 766 |
+
"grad_norm": 53.3145866394043,
|
| 767 |
+
"learning_rate": 2.710526315789474e-05,
|
| 768 |
+
"loss": 0.1186,
|
| 769 |
+
"step": 97
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.96,
|
| 773 |
+
"grad_norm": 13.420350074768066,
|
| 774 |
+
"learning_rate": 2.6842105263157896e-05,
|
| 775 |
+
"loss": 0.2332,
|
| 776 |
+
"step": 98
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.98,
|
| 780 |
+
"grad_norm": 13.3897123336792,
|
| 781 |
+
"learning_rate": 2.6578947368421052e-05,
|
| 782 |
+
"loss": 0.0759,
|
| 783 |
+
"step": 99
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.0,
|
| 787 |
+
"grad_norm": 1.8416231870651245,
|
| 788 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 789 |
+
"loss": 0.0076,
|
| 790 |
+
"step": 100
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.0,
|
| 794 |
+
"eval_accuracy": 0.993421052631579,
|
| 795 |
+
"eval_loss": 0.027568014338612556,
|
| 796 |
+
"eval_runtime": 7.0992,
|
| 797 |
+
"eval_samples_per_second": 42.822,
|
| 798 |
+
"eval_steps_per_second": 1.127,
|
| 799 |
+
"step": 100
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.02,
|
| 803 |
+
"grad_norm": 6.973387241363525,
|
| 804 |
+
"learning_rate": 2.605263157894737e-05,
|
| 805 |
+
"loss": 0.0255,
|
| 806 |
+
"step": 101
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.04,
|
| 810 |
+
"grad_norm": 0.33660009503364563,
|
| 811 |
+
"learning_rate": 2.578947368421053e-05,
|
| 812 |
+
"loss": 0.0012,
|
| 813 |
+
"step": 102
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.06,
|
| 817 |
+
"grad_norm": 11.269298553466797,
|
| 818 |
+
"learning_rate": 2.5526315789473688e-05,
|
| 819 |
+
"loss": 0.1097,
|
| 820 |
+
"step": 103
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.08,
|
| 824 |
+
"grad_norm": 7.602676868438721,
|
| 825 |
+
"learning_rate": 2.5263157894736845e-05,
|
| 826 |
+
"loss": 0.034,
|
| 827 |
+
"step": 104
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 2.1,
|
| 831 |
+
"grad_norm": 7.900285720825195,
|
| 832 |
+
"learning_rate": 2.5e-05,
|
| 833 |
+
"loss": 0.0399,
|
| 834 |
+
"step": 105
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.12,
|
| 838 |
+
"grad_norm": 4.07220983505249,
|
| 839 |
+
"learning_rate": 2.4736842105263158e-05,
|
| 840 |
+
"loss": 0.0055,
|
| 841 |
+
"step": 106
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.14,
|
| 845 |
+
"grad_norm": 0.09085848927497864,
|
| 846 |
+
"learning_rate": 2.4473684210526318e-05,
|
| 847 |
+
"loss": 0.0005,
|
| 848 |
+
"step": 107
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.16,
|
| 852 |
+
"grad_norm": 0.5666660666465759,
|
| 853 |
+
"learning_rate": 2.4210526315789474e-05,
|
| 854 |
+
"loss": 0.0029,
|
| 855 |
+
"step": 108
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 2.18,
|
| 859 |
+
"grad_norm": 12.90062427520752,
|
| 860 |
+
"learning_rate": 2.394736842105263e-05,
|
| 861 |
+
"loss": 0.051,
|
| 862 |
+
"step": 109
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 2.2,
|
| 866 |
+
"grad_norm": 0.24952377378940582,
|
| 867 |
+
"learning_rate": 2.368421052631579e-05,
|
| 868 |
+
"loss": 0.0007,
|
| 869 |
+
"step": 110
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 2.2,
|
| 873 |
+
"eval_accuracy": 0.9835526315789473,
|
| 874 |
+
"eval_loss": 0.05723079293966293,
|
| 875 |
+
"eval_runtime": 7.2757,
|
| 876 |
+
"eval_samples_per_second": 41.783,
|
| 877 |
+
"eval_steps_per_second": 1.1,
|
| 878 |
+
"step": 110
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.22,
|
| 882 |
+
"grad_norm": 2.654731512069702,
|
| 883 |
+
"learning_rate": 2.342105263157895e-05,
|
| 884 |
+
"loss": 0.0036,
|
| 885 |
+
"step": 111
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.24,
|
| 889 |
+
"grad_norm": 0.5114130973815918,
|
| 890 |
+
"learning_rate": 2.3157894736842107e-05,
|
| 891 |
+
"loss": 0.0012,
|
| 892 |
+
"step": 112
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 2.26,
|
| 896 |
+
"grad_norm": 9.55749797821045,
|
| 897 |
+
"learning_rate": 2.2894736842105263e-05,
|
| 898 |
+
"loss": 0.0195,
|
| 899 |
+
"step": 113
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.28,
|
| 903 |
+
"grad_norm": 11.102432250976562,
|
| 904 |
+
"learning_rate": 2.2631578947368423e-05,
|
| 905 |
+
"loss": 0.0204,
|
| 906 |
+
"step": 114
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 2.3,
|
| 910 |
+
"grad_norm": 2.684544324874878,
|
| 911 |
+
"learning_rate": 2.236842105263158e-05,
|
| 912 |
+
"loss": 0.0047,
|
| 913 |
+
"step": 115
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.32,
|
| 917 |
+
"grad_norm": 0.22236469388008118,
|
| 918 |
+
"learning_rate": 2.2105263157894736e-05,
|
| 919 |
+
"loss": 0.0008,
|
| 920 |
+
"step": 116
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 2.34,
|
| 924 |
+
"grad_norm": 8.926434516906738,
|
| 925 |
+
"learning_rate": 2.1842105263157896e-05,
|
| 926 |
+
"loss": 0.0223,
|
| 927 |
+
"step": 117
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.36,
|
| 931 |
+
"grad_norm": 3.3315141201019287,
|
| 932 |
+
"learning_rate": 2.1578947368421053e-05,
|
| 933 |
+
"loss": 0.0117,
|
| 934 |
+
"step": 118
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 2.38,
|
| 938 |
+
"grad_norm": 10.430353164672852,
|
| 939 |
+
"learning_rate": 2.1315789473684212e-05,
|
| 940 |
+
"loss": 0.0303,
|
| 941 |
+
"step": 119
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 2.4,
|
| 945 |
+
"grad_norm": 0.15297500789165497,
|
| 946 |
+
"learning_rate": 2.105263157894737e-05,
|
| 947 |
+
"loss": 0.0008,
|
| 948 |
+
"step": 120
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 2.4,
|
| 952 |
+
"eval_accuracy": 0.9967105263157895,
|
| 953 |
+
"eval_loss": 0.009093067608773708,
|
| 954 |
+
"eval_runtime": 6.8131,
|
| 955 |
+
"eval_samples_per_second": 44.62,
|
| 956 |
+
"eval_steps_per_second": 1.174,
|
| 957 |
+
"step": 120
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.42,
|
| 961 |
+
"grad_norm": 0.6949332356452942,
|
| 962 |
+
"learning_rate": 2.078947368421053e-05,
|
| 963 |
+
"loss": 0.0018,
|
| 964 |
+
"step": 121
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 2.44,
|
| 968 |
+
"grad_norm": 0.10584941506385803,
|
| 969 |
+
"learning_rate": 2.0526315789473685e-05,
|
| 970 |
+
"loss": 0.0006,
|
| 971 |
+
"step": 122
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.46,
|
| 975 |
+
"grad_norm": 2.782750368118286,
|
| 976 |
+
"learning_rate": 2.0263157894736842e-05,
|
| 977 |
+
"loss": 0.0064,
|
| 978 |
+
"step": 123
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 2.48,
|
| 982 |
+
"grad_norm": 0.1843283325433731,
|
| 983 |
+
"learning_rate": 2e-05,
|
| 984 |
+
"loss": 0.0007,
|
| 985 |
+
"step": 124
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 2.5,
|
| 989 |
+
"grad_norm": 19.077423095703125,
|
| 990 |
+
"learning_rate": 1.9736842105263158e-05,
|
| 991 |
+
"loss": 0.1095,
|
| 992 |
+
"step": 125
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.52,
|
| 996 |
+
"grad_norm": 9.631857872009277,
|
| 997 |
+
"learning_rate": 1.9473684210526315e-05,
|
| 998 |
+
"loss": 0.0286,
|
| 999 |
+
"step": 126
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.54,
|
| 1003 |
+
"grad_norm": 2.46711802482605,
|
| 1004 |
+
"learning_rate": 1.9210526315789474e-05,
|
| 1005 |
+
"loss": 0.0042,
|
| 1006 |
+
"step": 127
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 2.56,
|
| 1010 |
+
"grad_norm": 0.9471967220306396,
|
| 1011 |
+
"learning_rate": 1.8947368421052634e-05,
|
| 1012 |
+
"loss": 0.0038,
|
| 1013 |
+
"step": 128
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 2.58,
|
| 1017 |
+
"grad_norm": 0.1509537547826767,
|
| 1018 |
+
"learning_rate": 1.868421052631579e-05,
|
| 1019 |
+
"loss": 0.0009,
|
| 1020 |
+
"step": 129
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 2.6,
|
| 1024 |
+
"grad_norm": 16.95519256591797,
|
| 1025 |
+
"learning_rate": 1.8421052631578947e-05,
|
| 1026 |
+
"loss": 0.1277,
|
| 1027 |
+
"step": 130
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 2.6,
|
| 1031 |
+
"eval_accuracy": 0.9967105263157895,
|
| 1032 |
+
"eval_loss": 0.004848334938287735,
|
| 1033 |
+
"eval_runtime": 7.1393,
|
| 1034 |
+
"eval_samples_per_second": 42.581,
|
| 1035 |
+
"eval_steps_per_second": 1.121,
|
| 1036 |
+
"step": 130
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 2.62,
|
| 1040 |
+
"grad_norm": 0.3852033317089081,
|
| 1041 |
+
"learning_rate": 1.8157894736842107e-05,
|
| 1042 |
+
"loss": 0.0009,
|
| 1043 |
+
"step": 131
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.64,
|
| 1047 |
+
"grad_norm": 0.23005267977714539,
|
| 1048 |
+
"learning_rate": 1.7894736842105264e-05,
|
| 1049 |
+
"loss": 0.0006,
|
| 1050 |
+
"step": 132
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 2.66,
|
| 1054 |
+
"grad_norm": 2.9750654697418213,
|
| 1055 |
+
"learning_rate": 1.763157894736842e-05,
|
| 1056 |
+
"loss": 0.0096,
|
| 1057 |
+
"step": 133
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 2.68,
|
| 1061 |
+
"grad_norm": 6.330101490020752,
|
| 1062 |
+
"learning_rate": 1.736842105263158e-05,
|
| 1063 |
+
"loss": 0.0224,
|
| 1064 |
+
"step": 134
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.7,
|
| 1068 |
+
"grad_norm": 0.2537049949169159,
|
| 1069 |
+
"learning_rate": 1.7105263157894737e-05,
|
| 1070 |
+
"loss": 0.001,
|
| 1071 |
+
"step": 135
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 2.72,
|
| 1075 |
+
"grad_norm": 0.49432969093322754,
|
| 1076 |
+
"learning_rate": 1.6842105263157896e-05,
|
| 1077 |
+
"loss": 0.002,
|
| 1078 |
+
"step": 136
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 2.74,
|
| 1082 |
+
"grad_norm": 0.878919243812561,
|
| 1083 |
+
"learning_rate": 1.6578947368421053e-05,
|
| 1084 |
+
"loss": 0.0022,
|
| 1085 |
+
"step": 137
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 2.76,
|
| 1089 |
+
"grad_norm": 0.4538193941116333,
|
| 1090 |
+
"learning_rate": 1.6315789473684213e-05,
|
| 1091 |
+
"loss": 0.0016,
|
| 1092 |
+
"step": 138
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 2.78,
|
| 1096 |
+
"grad_norm": 13.346075057983398,
|
| 1097 |
+
"learning_rate": 1.605263157894737e-05,
|
| 1098 |
+
"loss": 0.0384,
|
| 1099 |
+
"step": 139
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 2.8,
|
| 1103 |
+
"grad_norm": 0.42253851890563965,
|
| 1104 |
+
"learning_rate": 1.5789473684210526e-05,
|
| 1105 |
+
"loss": 0.0007,
|
| 1106 |
+
"step": 140
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 2.8,
|
| 1110 |
+
"eval_accuracy": 0.9736842105263158,
|
| 1111 |
+
"eval_loss": 0.08703595399856567,
|
| 1112 |
+
"eval_runtime": 7.1737,
|
| 1113 |
+
"eval_samples_per_second": 42.377,
|
| 1114 |
+
"eval_steps_per_second": 1.115,
|
| 1115 |
+
"step": 140
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.82,
|
| 1119 |
+
"grad_norm": 8.999384880065918,
|
| 1120 |
+
"learning_rate": 1.5526315789473686e-05,
|
| 1121 |
+
"loss": 0.041,
|
| 1122 |
+
"step": 141
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.84,
|
| 1126 |
+
"grad_norm": 0.34943702816963196,
|
| 1127 |
+
"learning_rate": 1.5263157894736842e-05,
|
| 1128 |
+
"loss": 0.0008,
|
| 1129 |
+
"step": 142
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.86,
|
| 1133 |
+
"grad_norm": 41.58428192138672,
|
| 1134 |
+
"learning_rate": 1.5e-05,
|
| 1135 |
+
"loss": 0.0352,
|
| 1136 |
+
"step": 143
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.88,
|
| 1140 |
+
"grad_norm": 1.358328104019165,
|
| 1141 |
+
"learning_rate": 1.4736842105263157e-05,
|
| 1142 |
+
"loss": 0.004,
|
| 1143 |
+
"step": 144
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.9,
|
| 1147 |
+
"grad_norm": 26.084732055664062,
|
| 1148 |
+
"learning_rate": 1.4473684210526317e-05,
|
| 1149 |
+
"loss": 0.1456,
|
| 1150 |
+
"step": 145
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.92,
|
| 1154 |
+
"grad_norm": 4.426596164703369,
|
| 1155 |
+
"learning_rate": 1.4210526315789475e-05,
|
| 1156 |
+
"loss": 0.0056,
|
| 1157 |
+
"step": 146
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.94,
|
| 1161 |
+
"grad_norm": 7.729344367980957,
|
| 1162 |
+
"learning_rate": 1.3947368421052631e-05,
|
| 1163 |
+
"loss": 0.0536,
|
| 1164 |
+
"step": 147
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.96,
|
| 1168 |
+
"grad_norm": 0.49353259801864624,
|
| 1169 |
+
"learning_rate": 1.3684210526315791e-05,
|
| 1170 |
+
"loss": 0.0035,
|
| 1171 |
+
"step": 148
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.98,
|
| 1175 |
+
"grad_norm": 15.301690101623535,
|
| 1176 |
+
"learning_rate": 1.3421052631578948e-05,
|
| 1177 |
+
"loss": 0.0138,
|
| 1178 |
+
"step": 149
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.0,
|
| 1182 |
+
"grad_norm": 0.520106852054596,
|
| 1183 |
+
"learning_rate": 1.3157894736842106e-05,
|
| 1184 |
+
"loss": 0.0008,
|
| 1185 |
+
"step": 150
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.0,
|
| 1189 |
+
"eval_accuracy": 0.993421052631579,
|
| 1190 |
+
"eval_loss": 0.005907059647142887,
|
| 1191 |
+
"eval_runtime": 7.2044,
|
| 1192 |
+
"eval_samples_per_second": 42.197,
|
| 1193 |
+
"eval_steps_per_second": 1.11,
|
| 1194 |
+
"step": 150
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 3.02,
|
| 1198 |
+
"grad_norm": 0.7136516571044922,
|
| 1199 |
+
"learning_rate": 1.2894736842105264e-05,
|
| 1200 |
+
"loss": 0.0028,
|
| 1201 |
+
"step": 151
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 3.04,
|
| 1205 |
+
"grad_norm": 0.28995925188064575,
|
| 1206 |
+
"learning_rate": 1.2631578947368422e-05,
|
| 1207 |
+
"loss": 0.0006,
|
| 1208 |
+
"step": 152
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 3.06,
|
| 1212 |
+
"grad_norm": 9.869847297668457,
|
| 1213 |
+
"learning_rate": 1.2368421052631579e-05,
|
| 1214 |
+
"loss": 0.0342,
|
| 1215 |
+
"step": 153
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 3.08,
|
| 1219 |
+
"grad_norm": 0.20523004233837128,
|
| 1220 |
+
"learning_rate": 1.2105263157894737e-05,
|
| 1221 |
+
"loss": 0.0005,
|
| 1222 |
+
"step": 154
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 3.1,
|
| 1226 |
+
"grad_norm": 8.038060188293457,
|
| 1227 |
+
"learning_rate": 1.1842105263157895e-05,
|
| 1228 |
+
"loss": 0.0178,
|
| 1229 |
+
"step": 155
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 3.12,
|
| 1233 |
+
"grad_norm": 1.2011505365371704,
|
| 1234 |
+
"learning_rate": 1.1578947368421053e-05,
|
| 1235 |
+
"loss": 0.002,
|
| 1236 |
+
"step": 156
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 3.14,
|
| 1240 |
+
"grad_norm": 1.0488544702529907,
|
| 1241 |
+
"learning_rate": 1.1315789473684212e-05,
|
| 1242 |
+
"loss": 0.0025,
|
| 1243 |
+
"step": 157
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 3.16,
|
| 1247 |
+
"grad_norm": 0.24755439162254333,
|
| 1248 |
+
"learning_rate": 1.1052631578947368e-05,
|
| 1249 |
+
"loss": 0.0005,
|
| 1250 |
+
"step": 158
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 3.18,
|
| 1254 |
+
"grad_norm": 0.37720346450805664,
|
| 1255 |
+
"learning_rate": 1.0789473684210526e-05,
|
| 1256 |
+
"loss": 0.0007,
|
| 1257 |
+
"step": 159
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 3.2,
|
| 1261 |
+
"grad_norm": 0.054420772939920425,
|
| 1262 |
+
"learning_rate": 1.0526315789473684e-05,
|
| 1263 |
+
"loss": 0.0003,
|
| 1264 |
+
"step": 160
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 3.2,
|
| 1268 |
+
"eval_accuracy": 0.9967105263157895,
|
| 1269 |
+
"eval_loss": 0.012113516218960285,
|
| 1270 |
+
"eval_runtime": 7.4779,
|
| 1271 |
+
"eval_samples_per_second": 40.653,
|
| 1272 |
+
"eval_steps_per_second": 1.07,
|
| 1273 |
+
"step": 160
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 3.22,
|
| 1277 |
+
"grad_norm": 3.170163631439209,
|
| 1278 |
+
"learning_rate": 1.0263157894736843e-05,
|
| 1279 |
+
"loss": 0.0048,
|
| 1280 |
+
"step": 161
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 3.24,
|
| 1284 |
+
"grad_norm": 9.659165382385254,
|
| 1285 |
+
"learning_rate": 1e-05,
|
| 1286 |
+
"loss": 0.006,
|
| 1287 |
+
"step": 162
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 3.26,
|
| 1291 |
+
"grad_norm": 0.07884739339351654,
|
| 1292 |
+
"learning_rate": 9.736842105263157e-06,
|
| 1293 |
+
"loss": 0.0003,
|
| 1294 |
+
"step": 163
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 3.28,
|
| 1298 |
+
"grad_norm": 1.1162643432617188,
|
| 1299 |
+
"learning_rate": 9.473684210526317e-06,
|
| 1300 |
+
"loss": 0.0019,
|
| 1301 |
+
"step": 164
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 3.3,
|
| 1305 |
+
"grad_norm": 0.3206580877304077,
|
| 1306 |
+
"learning_rate": 9.210526315789474e-06,
|
| 1307 |
+
"loss": 0.001,
|
| 1308 |
+
"step": 165
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 3.32,
|
| 1312 |
+
"grad_norm": 15.26516056060791,
|
| 1313 |
+
"learning_rate": 8.947368421052632e-06,
|
| 1314 |
+
"loss": 0.0467,
|
| 1315 |
+
"step": 166
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 3.34,
|
| 1319 |
+
"grad_norm": 0.6477659940719604,
|
| 1320 |
+
"learning_rate": 8.68421052631579e-06,
|
| 1321 |
+
"loss": 0.002,
|
| 1322 |
+
"step": 167
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 3.36,
|
| 1326 |
+
"grad_norm": 16.165756225585938,
|
| 1327 |
+
"learning_rate": 8.421052631578948e-06,
|
| 1328 |
+
"loss": 0.0486,
|
| 1329 |
+
"step": 168
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 3.38,
|
| 1333 |
+
"grad_norm": 0.3942013382911682,
|
| 1334 |
+
"learning_rate": 8.157894736842106e-06,
|
| 1335 |
+
"loss": 0.001,
|
| 1336 |
+
"step": 169
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 3.4,
|
| 1340 |
+
"grad_norm": 0.2746890187263489,
|
| 1341 |
+
"learning_rate": 7.894736842105263e-06,
|
| 1342 |
+
"loss": 0.0008,
|
| 1343 |
+
"step": 170
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 3.4,
|
| 1347 |
+
"eval_accuracy": 0.993421052631579,
|
| 1348 |
+
"eval_loss": 0.02530418336391449,
|
| 1349 |
+
"eval_runtime": 6.8199,
|
| 1350 |
+
"eval_samples_per_second": 44.575,
|
| 1351 |
+
"eval_steps_per_second": 1.173,
|
| 1352 |
+
"step": 170
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 3.42,
|
| 1356 |
+
"grad_norm": 0.1466667354106903,
|
| 1357 |
+
"learning_rate": 7.631578947368421e-06,
|
| 1358 |
+
"loss": 0.0006,
|
| 1359 |
+
"step": 171
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 3.44,
|
| 1363 |
+
"grad_norm": 0.35398414731025696,
|
| 1364 |
+
"learning_rate": 7.3684210526315784e-06,
|
| 1365 |
+
"loss": 0.0008,
|
| 1366 |
+
"step": 172
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 3.46,
|
| 1370 |
+
"grad_norm": 0.03358185663819313,
|
| 1371 |
+
"learning_rate": 7.1052631578947375e-06,
|
| 1372 |
+
"loss": 0.0001,
|
| 1373 |
+
"step": 173
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 3.48,
|
| 1377 |
+
"grad_norm": 0.04026187211275101,
|
| 1378 |
+
"learning_rate": 6.842105263157896e-06,
|
| 1379 |
+
"loss": 0.0002,
|
| 1380 |
+
"step": 174
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 3.5,
|
| 1384 |
+
"grad_norm": 0.10377287119626999,
|
| 1385 |
+
"learning_rate": 6.578947368421053e-06,
|
| 1386 |
+
"loss": 0.0008,
|
| 1387 |
+
"step": 175
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 3.52,
|
| 1391 |
+
"grad_norm": 3.1232450008392334,
|
| 1392 |
+
"learning_rate": 6.315789473684211e-06,
|
| 1393 |
+
"loss": 0.0062,
|
| 1394 |
+
"step": 176
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 3.54,
|
| 1398 |
+
"grad_norm": 0.0659070760011673,
|
| 1399 |
+
"learning_rate": 6.0526315789473685e-06,
|
| 1400 |
+
"loss": 0.0002,
|
| 1401 |
+
"step": 177
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 3.56,
|
| 1405 |
+
"grad_norm": 0.4386169910430908,
|
| 1406 |
+
"learning_rate": 5.789473684210527e-06,
|
| 1407 |
+
"loss": 0.0007,
|
| 1408 |
+
"step": 178
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 3.58,
|
| 1412 |
+
"grad_norm": 0.09403068572282791,
|
| 1413 |
+
"learning_rate": 5.526315789473684e-06,
|
| 1414 |
+
"loss": 0.0001,
|
| 1415 |
+
"step": 179
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 3.6,
|
| 1419 |
+
"grad_norm": 0.011226998642086983,
|
| 1420 |
+
"learning_rate": 5.263157894736842e-06,
|
| 1421 |
+
"loss": 0.0,
|
| 1422 |
+
"step": 180
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 3.6,
|
| 1426 |
+
"eval_accuracy": 0.9967105263157895,
|
| 1427 |
+
"eval_loss": 0.012991190887987614,
|
| 1428 |
+
"eval_runtime": 7.2339,
|
| 1429 |
+
"eval_samples_per_second": 42.024,
|
| 1430 |
+
"eval_steps_per_second": 1.106,
|
| 1431 |
+
"step": 180
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 3.62,
|
| 1435 |
+
"grad_norm": 0.3960583209991455,
|
| 1436 |
+
"learning_rate": 5e-06,
|
| 1437 |
+
"loss": 0.0013,
|
| 1438 |
+
"step": 181
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 3.64,
|
| 1442 |
+
"grad_norm": 0.028123315423727036,
|
| 1443 |
+
"learning_rate": 4.736842105263159e-06,
|
| 1444 |
+
"loss": 0.0001,
|
| 1445 |
+
"step": 182
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 3.66,
|
| 1449 |
+
"grad_norm": 3.7100679874420166,
|
| 1450 |
+
"learning_rate": 4.473684210526316e-06,
|
| 1451 |
+
"loss": 0.0062,
|
| 1452 |
+
"step": 183
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 3.68,
|
| 1456 |
+
"grad_norm": 0.027926042675971985,
|
| 1457 |
+
"learning_rate": 4.210526315789474e-06,
|
| 1458 |
+
"loss": 0.0001,
|
| 1459 |
+
"step": 184
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 3.7,
|
| 1463 |
+
"grad_norm": 1.727767825126648,
|
| 1464 |
+
"learning_rate": 3.9473684210526315e-06,
|
| 1465 |
+
"loss": 0.0022,
|
| 1466 |
+
"step": 185
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 3.72,
|
| 1470 |
+
"grad_norm": 1.4793553352355957,
|
| 1471 |
+
"learning_rate": 3.6842105263157892e-06,
|
| 1472 |
+
"loss": 0.0011,
|
| 1473 |
+
"step": 186
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 3.74,
|
| 1477 |
+
"grad_norm": 2.847693681716919,
|
| 1478 |
+
"learning_rate": 3.421052631578948e-06,
|
| 1479 |
+
"loss": 0.004,
|
| 1480 |
+
"step": 187
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 3.76,
|
| 1484 |
+
"grad_norm": 0.017291821539402008,
|
| 1485 |
+
"learning_rate": 3.1578947368421056e-06,
|
| 1486 |
+
"loss": 0.0001,
|
| 1487 |
+
"step": 188
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 3.78,
|
| 1491 |
+
"grad_norm": 0.6957990527153015,
|
| 1492 |
+
"learning_rate": 2.8947368421052634e-06,
|
| 1493 |
+
"loss": 0.001,
|
| 1494 |
+
"step": 189
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.8,
|
| 1498 |
+
"grad_norm": 0.032589249312877655,
|
| 1499 |
+
"learning_rate": 2.631578947368421e-06,
|
| 1500 |
+
"loss": 0.0001,
|
| 1501 |
+
"step": 190
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 3.8,
|
| 1505 |
+
"eval_accuracy": 0.993421052631579,
|
| 1506 |
+
"eval_loss": 0.009236541576683521,
|
| 1507 |
+
"eval_runtime": 6.8219,
|
| 1508 |
+
"eval_samples_per_second": 44.563,
|
| 1509 |
+
"eval_steps_per_second": 1.173,
|
| 1510 |
+
"step": 190
|
| 1511 |
+
}
|
| 1512 |
+
],
|
| 1513 |
+
"logging_steps": 1,
|
| 1514 |
+
"max_steps": 200,
|
| 1515 |
+
"num_input_tokens_seen": 0,
|
| 1516 |
+
"num_train_epochs": 4,
|
| 1517 |
+
"save_steps": 10,
|
| 1518 |
+
"total_flos": 5.904904527684772e+17,
|
| 1519 |
+
"train_batch_size": 40,
|
| 1520 |
+
"trial_name": null,
|
| 1521 |
+
"trial_params": null
|
| 1522 |
+
}
|
checkpoint-190/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d12326b817ccf86422554696b6e15d756b4174e67e313ecfbe14aadf3bb73244
|
| 3 |
+
size 4856
|
checkpoint-200/config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Organika/sdxl-detector",
|
| 3 |
+
"_num_labels": 2,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"SwinForImageClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.0,
|
| 8 |
+
"depths": [
|
| 9 |
+
2,
|
| 10 |
+
2,
|
| 11 |
+
18,
|
| 12 |
+
2
|
| 13 |
+
],
|
| 14 |
+
"drop_path_rate": 0.1,
|
| 15 |
+
"embed_dim": 128,
|
| 16 |
+
"encoder_stride": 32,
|
| 17 |
+
"hidden_act": "gelu",
|
| 18 |
+
"hidden_dropout_prob": 0.0,
|
| 19 |
+
"hidden_size": 1024,
|
| 20 |
+
"id2label": {
|
| 21 |
+
"0": "artificial",
|
| 22 |
+
"1": "human"
|
| 23 |
+
},
|
| 24 |
+
"image_size": 224,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"label2id": {
|
| 27 |
+
"artificial": 0,
|
| 28 |
+
"human": 1
|
| 29 |
+
},
|
| 30 |
+
"layer_norm_eps": 1e-05,
|
| 31 |
+
"max_length": 128,
|
| 32 |
+
"mlp_ratio": 4.0,
|
| 33 |
+
"model_type": "swin",
|
| 34 |
+
"num_channels": 3,
|
| 35 |
+
"num_heads": [
|
| 36 |
+
4,
|
| 37 |
+
8,
|
| 38 |
+
16,
|
| 39 |
+
32
|
| 40 |
+
],
|
| 41 |
+
"num_layers": 4,
|
| 42 |
+
"out_features": [
|
| 43 |
+
"stage4"
|
| 44 |
+
],
|
| 45 |
+
"out_indices": [
|
| 46 |
+
4
|
| 47 |
+
],
|
| 48 |
+
"padding": "max_length",
|
| 49 |
+
"patch_size": 4,
|
| 50 |
+
"path_norm": true,
|
| 51 |
+
"problem_type": "single_label_classification",
|
| 52 |
+
"qkv_bias": true,
|
| 53 |
+
"stage_names": [
|
| 54 |
+
"stem",
|
| 55 |
+
"stage1",
|
| 56 |
+
"stage2",
|
| 57 |
+
"stage3",
|
| 58 |
+
"stage4"
|
| 59 |
+
],
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.39.3",
|
| 62 |
+
"use_absolute_embeddings": false,
|
| 63 |
+
"window_size": 7
|
| 64 |
+
}
|
checkpoint-200/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d1722542eb8997bf4925c0254af0bf6c6ac2c9b3499879ce5aaa0b4d3f48f83
|
| 3 |
+
size 347498816
|
checkpoint-200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4182967f6667786b57d0f91b8f1268d3194964249208ad46de82ad989737fc71
|
| 3 |
+
size 694317645
|
checkpoint-200/preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.485,
|
| 21 |
+
0.456,
|
| 22 |
+
0.406
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.229,
|
| 27 |
+
0.224,
|
| 28 |
+
0.225
|
| 29 |
+
],
|
| 30 |
+
"resample": 3,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
checkpoint-200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c2bf14f56667191904a767265aae5cce753dc5f2ae8bc6452ff41d6d9ac99fc0
|
| 3 |
+
size 14244
|
checkpoint-200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a27b052646dcb561cbd68156c30bd466ce59bda64cf3c8eba9c3c1113af9827c
|
| 3 |
+
size 1064
|
checkpoint-200/trainer_state.json
ADDED
|
@@ -0,0 +1,1601 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.004848334938287735,
|
| 3 |
+
"best_model_checkpoint": "ai_vs_real_image_detection/checkpoint-130",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 52.89345169067383,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 2.2763,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04,
|
| 20 |
+
"grad_norm": 62.66461944580078,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.8254,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06,
|
| 27 |
+
"grad_norm": 66.1015625,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 1.8906,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08,
|
| 34 |
+
"grad_norm": 56.700233459472656,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 1.3529,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1,
|
| 41 |
+
"grad_norm": 66.10917663574219,
|
| 42 |
+
"learning_rate": 2.5e-05,
|
| 43 |
+
"loss": 1.4672,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12,
|
| 48 |
+
"grad_norm": 52.68284225463867,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 1.0578,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14,
|
| 55 |
+
"grad_norm": 42.4580078125,
|
| 56 |
+
"learning_rate": 3.5e-05,
|
| 57 |
+
"loss": 1.1221,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.16,
|
| 62 |
+
"grad_norm": 64.87982940673828,
|
| 63 |
+
"learning_rate": 4e-05,
|
| 64 |
+
"loss": 0.9313,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.18,
|
| 69 |
+
"grad_norm": 56.227081298828125,
|
| 70 |
+
"learning_rate": 4.5e-05,
|
| 71 |
+
"loss": 0.644,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2,
|
| 76 |
+
"grad_norm": 13.719300270080566,
|
| 77 |
+
"learning_rate": 5e-05,
|
| 78 |
+
"loss": 0.4305,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2,
|
| 83 |
+
"eval_accuracy": 0.9144736842105263,
|
| 84 |
+
"eval_loss": 0.19980163872241974,
|
| 85 |
+
"eval_runtime": 6.726,
|
| 86 |
+
"eval_samples_per_second": 45.198,
|
| 87 |
+
"eval_steps_per_second": 1.189,
|
| 88 |
+
"step": 10
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.22,
|
| 92 |
+
"grad_norm": 16.044822692871094,
|
| 93 |
+
"learning_rate": 4.973684210526316e-05,
|
| 94 |
+
"loss": 0.3892,
|
| 95 |
+
"step": 11
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.24,
|
| 99 |
+
"grad_norm": 15.53343391418457,
|
| 100 |
+
"learning_rate": 4.9473684210526315e-05,
|
| 101 |
+
"loss": 0.3683,
|
| 102 |
+
"step": 12
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.26,
|
| 106 |
+
"grad_norm": 13.435264587402344,
|
| 107 |
+
"learning_rate": 4.921052631578947e-05,
|
| 108 |
+
"loss": 0.3594,
|
| 109 |
+
"step": 13
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.28,
|
| 113 |
+
"grad_norm": 17.45833969116211,
|
| 114 |
+
"learning_rate": 4.8947368421052635e-05,
|
| 115 |
+
"loss": 0.5304,
|
| 116 |
+
"step": 14
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.3,
|
| 120 |
+
"grad_norm": 14.26673412322998,
|
| 121 |
+
"learning_rate": 4.868421052631579e-05,
|
| 122 |
+
"loss": 0.4804,
|
| 123 |
+
"step": 15
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.32,
|
| 127 |
+
"grad_norm": 9.26322078704834,
|
| 128 |
+
"learning_rate": 4.842105263157895e-05,
|
| 129 |
+
"loss": 0.3633,
|
| 130 |
+
"step": 16
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.34,
|
| 134 |
+
"grad_norm": 8.355168342590332,
|
| 135 |
+
"learning_rate": 4.8157894736842105e-05,
|
| 136 |
+
"loss": 0.4711,
|
| 137 |
+
"step": 17
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.36,
|
| 141 |
+
"grad_norm": 8.733236312866211,
|
| 142 |
+
"learning_rate": 4.789473684210526e-05,
|
| 143 |
+
"loss": 0.3954,
|
| 144 |
+
"step": 18
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.38,
|
| 148 |
+
"grad_norm": 5.385647296905518,
|
| 149 |
+
"learning_rate": 4.7631578947368424e-05,
|
| 150 |
+
"loss": 0.1943,
|
| 151 |
+
"step": 19
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.4,
|
| 155 |
+
"grad_norm": 9.892962455749512,
|
| 156 |
+
"learning_rate": 4.736842105263158e-05,
|
| 157 |
+
"loss": 0.3209,
|
| 158 |
+
"step": 20
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.4,
|
| 162 |
+
"eval_accuracy": 0.8881578947368421,
|
| 163 |
+
"eval_loss": 0.24087880551815033,
|
| 164 |
+
"eval_runtime": 6.8213,
|
| 165 |
+
"eval_samples_per_second": 44.566,
|
| 166 |
+
"eval_steps_per_second": 1.173,
|
| 167 |
+
"step": 20
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.42,
|
| 171 |
+
"grad_norm": 5.46142578125,
|
| 172 |
+
"learning_rate": 4.7105263157894744e-05,
|
| 173 |
+
"loss": 0.1505,
|
| 174 |
+
"step": 21
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.44,
|
| 178 |
+
"grad_norm": 9.863460540771484,
|
| 179 |
+
"learning_rate": 4.68421052631579e-05,
|
| 180 |
+
"loss": 0.2481,
|
| 181 |
+
"step": 22
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.46,
|
| 185 |
+
"grad_norm": 7.595032691955566,
|
| 186 |
+
"learning_rate": 4.657894736842106e-05,
|
| 187 |
+
"loss": 0.2001,
|
| 188 |
+
"step": 23
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.48,
|
| 192 |
+
"grad_norm": 17.375865936279297,
|
| 193 |
+
"learning_rate": 4.6315789473684214e-05,
|
| 194 |
+
"loss": 0.336,
|
| 195 |
+
"step": 24
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.5,
|
| 199 |
+
"grad_norm": 5.597734451293945,
|
| 200 |
+
"learning_rate": 4.605263157894737e-05,
|
| 201 |
+
"loss": 0.1395,
|
| 202 |
+
"step": 25
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.52,
|
| 206 |
+
"grad_norm": 7.56685209274292,
|
| 207 |
+
"learning_rate": 4.5789473684210527e-05,
|
| 208 |
+
"loss": 0.1739,
|
| 209 |
+
"step": 26
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.54,
|
| 213 |
+
"grad_norm": 13.769929885864258,
|
| 214 |
+
"learning_rate": 4.552631578947369e-05,
|
| 215 |
+
"loss": 0.1959,
|
| 216 |
+
"step": 27
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.56,
|
| 220 |
+
"grad_norm": 10.677892684936523,
|
| 221 |
+
"learning_rate": 4.5263157894736846e-05,
|
| 222 |
+
"loss": 0.2015,
|
| 223 |
+
"step": 28
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.58,
|
| 227 |
+
"grad_norm": 3.645374059677124,
|
| 228 |
+
"learning_rate": 4.5e-05,
|
| 229 |
+
"loss": 0.0576,
|
| 230 |
+
"step": 29
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.6,
|
| 234 |
+
"grad_norm": 12.538244247436523,
|
| 235 |
+
"learning_rate": 4.473684210526316e-05,
|
| 236 |
+
"loss": 0.324,
|
| 237 |
+
"step": 30
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.6,
|
| 241 |
+
"eval_accuracy": 0.9802631578947368,
|
| 242 |
+
"eval_loss": 0.056003469973802567,
|
| 243 |
+
"eval_runtime": 6.4198,
|
| 244 |
+
"eval_samples_per_second": 47.353,
|
| 245 |
+
"eval_steps_per_second": 1.246,
|
| 246 |
+
"step": 30
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.62,
|
| 250 |
+
"grad_norm": 8.700516700744629,
|
| 251 |
+
"learning_rate": 4.4473684210526316e-05,
|
| 252 |
+
"loss": 0.1199,
|
| 253 |
+
"step": 31
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.64,
|
| 257 |
+
"grad_norm": 6.926602840423584,
|
| 258 |
+
"learning_rate": 4.421052631578947e-05,
|
| 259 |
+
"loss": 0.1471,
|
| 260 |
+
"step": 32
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.66,
|
| 264 |
+
"grad_norm": 8.461554527282715,
|
| 265 |
+
"learning_rate": 4.394736842105263e-05,
|
| 266 |
+
"loss": 0.0649,
|
| 267 |
+
"step": 33
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.68,
|
| 271 |
+
"grad_norm": 7.129871845245361,
|
| 272 |
+
"learning_rate": 4.368421052631579e-05,
|
| 273 |
+
"loss": 0.1167,
|
| 274 |
+
"step": 34
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.7,
|
| 278 |
+
"grad_norm": 20.945388793945312,
|
| 279 |
+
"learning_rate": 4.342105263157895e-05,
|
| 280 |
+
"loss": 0.3279,
|
| 281 |
+
"step": 35
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.72,
|
| 285 |
+
"grad_norm": 4.466580390930176,
|
| 286 |
+
"learning_rate": 4.3157894736842105e-05,
|
| 287 |
+
"loss": 0.0517,
|
| 288 |
+
"step": 36
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.74,
|
| 292 |
+
"grad_norm": 9.97375202178955,
|
| 293 |
+
"learning_rate": 4.289473684210527e-05,
|
| 294 |
+
"loss": 0.1241,
|
| 295 |
+
"step": 37
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.76,
|
| 299 |
+
"grad_norm": 6.459863662719727,
|
| 300 |
+
"learning_rate": 4.2631578947368425e-05,
|
| 301 |
+
"loss": 0.0859,
|
| 302 |
+
"step": 38
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.78,
|
| 306 |
+
"grad_norm": 11.704880714416504,
|
| 307 |
+
"learning_rate": 4.236842105263158e-05,
|
| 308 |
+
"loss": 0.1252,
|
| 309 |
+
"step": 39
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.8,
|
| 313 |
+
"grad_norm": 7.90261697769165,
|
| 314 |
+
"learning_rate": 4.210526315789474e-05,
|
| 315 |
+
"loss": 0.0541,
|
| 316 |
+
"step": 40
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.8,
|
| 320 |
+
"eval_accuracy": 0.9671052631578947,
|
| 321 |
+
"eval_loss": 0.06965293735265732,
|
| 322 |
+
"eval_runtime": 6.6281,
|
| 323 |
+
"eval_samples_per_second": 45.865,
|
| 324 |
+
"eval_steps_per_second": 1.207,
|
| 325 |
+
"step": 40
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.82,
|
| 329 |
+
"grad_norm": 2.8429932594299316,
|
| 330 |
+
"learning_rate": 4.18421052631579e-05,
|
| 331 |
+
"loss": 0.0325,
|
| 332 |
+
"step": 41
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.84,
|
| 336 |
+
"grad_norm": 10.533605575561523,
|
| 337 |
+
"learning_rate": 4.157894736842106e-05,
|
| 338 |
+
"loss": 0.1836,
|
| 339 |
+
"step": 42
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.86,
|
| 343 |
+
"grad_norm": 12.08105754852295,
|
| 344 |
+
"learning_rate": 4.1315789473684214e-05,
|
| 345 |
+
"loss": 0.224,
|
| 346 |
+
"step": 43
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.88,
|
| 350 |
+
"grad_norm": 9.006488800048828,
|
| 351 |
+
"learning_rate": 4.105263157894737e-05,
|
| 352 |
+
"loss": 0.2139,
|
| 353 |
+
"step": 44
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9,
|
| 357 |
+
"grad_norm": 7.3875956535339355,
|
| 358 |
+
"learning_rate": 4.078947368421053e-05,
|
| 359 |
+
"loss": 0.088,
|
| 360 |
+
"step": 45
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.92,
|
| 364 |
+
"grad_norm": 13.019746780395508,
|
| 365 |
+
"learning_rate": 4.0526315789473684e-05,
|
| 366 |
+
"loss": 0.2363,
|
| 367 |
+
"step": 46
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.94,
|
| 371 |
+
"grad_norm": 37.95082473754883,
|
| 372 |
+
"learning_rate": 4.026315789473684e-05,
|
| 373 |
+
"loss": 0.3728,
|
| 374 |
+
"step": 47
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.96,
|
| 378 |
+
"grad_norm": 13.99156665802002,
|
| 379 |
+
"learning_rate": 4e-05,
|
| 380 |
+
"loss": 0.1376,
|
| 381 |
+
"step": 48
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.98,
|
| 385 |
+
"grad_norm": 5.063250541687012,
|
| 386 |
+
"learning_rate": 3.973684210526316e-05,
|
| 387 |
+
"loss": 0.0832,
|
| 388 |
+
"step": 49
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"grad_norm": 13.046541213989258,
|
| 393 |
+
"learning_rate": 3.9473684210526316e-05,
|
| 394 |
+
"loss": 0.1096,
|
| 395 |
+
"step": 50
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.0,
|
| 399 |
+
"eval_accuracy": 0.9572368421052632,
|
| 400 |
+
"eval_loss": 0.12372467666864395,
|
| 401 |
+
"eval_runtime": 6.8765,
|
| 402 |
+
"eval_samples_per_second": 44.209,
|
| 403 |
+
"eval_steps_per_second": 1.163,
|
| 404 |
+
"step": 50
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.02,
|
| 408 |
+
"grad_norm": 4.865447998046875,
|
| 409 |
+
"learning_rate": 3.921052631578947e-05,
|
| 410 |
+
"loss": 0.0366,
|
| 411 |
+
"step": 51
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.04,
|
| 415 |
+
"grad_norm": 12.430005073547363,
|
| 416 |
+
"learning_rate": 3.894736842105263e-05,
|
| 417 |
+
"loss": 0.1301,
|
| 418 |
+
"step": 52
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 1.06,
|
| 422 |
+
"grad_norm": 2.1291539669036865,
|
| 423 |
+
"learning_rate": 3.868421052631579e-05,
|
| 424 |
+
"loss": 0.0168,
|
| 425 |
+
"step": 53
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.08,
|
| 429 |
+
"grad_norm": 1.6922448873519897,
|
| 430 |
+
"learning_rate": 3.842105263157895e-05,
|
| 431 |
+
"loss": 0.0074,
|
| 432 |
+
"step": 54
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.1,
|
| 436 |
+
"grad_norm": 5.851077556610107,
|
| 437 |
+
"learning_rate": 3.815789473684211e-05,
|
| 438 |
+
"loss": 0.0216,
|
| 439 |
+
"step": 55
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.12,
|
| 443 |
+
"grad_norm": 21.35593605041504,
|
| 444 |
+
"learning_rate": 3.789473684210527e-05,
|
| 445 |
+
"loss": 0.1171,
|
| 446 |
+
"step": 56
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.14,
|
| 450 |
+
"grad_norm": 9.791412353515625,
|
| 451 |
+
"learning_rate": 3.7631578947368425e-05,
|
| 452 |
+
"loss": 0.1688,
|
| 453 |
+
"step": 57
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.16,
|
| 457 |
+
"grad_norm": 3.2542812824249268,
|
| 458 |
+
"learning_rate": 3.736842105263158e-05,
|
| 459 |
+
"loss": 0.0236,
|
| 460 |
+
"step": 58
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.18,
|
| 464 |
+
"grad_norm": 9.557928085327148,
|
| 465 |
+
"learning_rate": 3.710526315789474e-05,
|
| 466 |
+
"loss": 0.1782,
|
| 467 |
+
"step": 59
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.2,
|
| 471 |
+
"grad_norm": 12.514122009277344,
|
| 472 |
+
"learning_rate": 3.6842105263157895e-05,
|
| 473 |
+
"loss": 0.0905,
|
| 474 |
+
"step": 60
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.2,
|
| 478 |
+
"eval_accuracy": 0.9967105263157895,
|
| 479 |
+
"eval_loss": 0.027645185589790344,
|
| 480 |
+
"eval_runtime": 7.1325,
|
| 481 |
+
"eval_samples_per_second": 42.622,
|
| 482 |
+
"eval_steps_per_second": 1.122,
|
| 483 |
+
"step": 60
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.22,
|
| 487 |
+
"grad_norm": 4.434255599975586,
|
| 488 |
+
"learning_rate": 3.657894736842106e-05,
|
| 489 |
+
"loss": 0.0276,
|
| 490 |
+
"step": 61
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.24,
|
| 494 |
+
"grad_norm": 6.304673671722412,
|
| 495 |
+
"learning_rate": 3.6315789473684214e-05,
|
| 496 |
+
"loss": 0.0508,
|
| 497 |
+
"step": 62
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.26,
|
| 501 |
+
"grad_norm": 3.844200372695923,
|
| 502 |
+
"learning_rate": 3.605263157894737e-05,
|
| 503 |
+
"loss": 0.0199,
|
| 504 |
+
"step": 63
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.28,
|
| 508 |
+
"grad_norm": 0.8185235261917114,
|
| 509 |
+
"learning_rate": 3.578947368421053e-05,
|
| 510 |
+
"loss": 0.0069,
|
| 511 |
+
"step": 64
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.3,
|
| 515 |
+
"grad_norm": 6.909903049468994,
|
| 516 |
+
"learning_rate": 3.5526315789473684e-05,
|
| 517 |
+
"loss": 0.0597,
|
| 518 |
+
"step": 65
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.32,
|
| 522 |
+
"grad_norm": 3.5652379989624023,
|
| 523 |
+
"learning_rate": 3.526315789473684e-05,
|
| 524 |
+
"loss": 0.0141,
|
| 525 |
+
"step": 66
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.34,
|
| 529 |
+
"grad_norm": 11.72857666015625,
|
| 530 |
+
"learning_rate": 3.5e-05,
|
| 531 |
+
"loss": 0.1865,
|
| 532 |
+
"step": 67
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.36,
|
| 536 |
+
"grad_norm": 10.541884422302246,
|
| 537 |
+
"learning_rate": 3.473684210526316e-05,
|
| 538 |
+
"loss": 0.0606,
|
| 539 |
+
"step": 68
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.38,
|
| 543 |
+
"grad_norm": 5.210526943206787,
|
| 544 |
+
"learning_rate": 3.447368421052632e-05,
|
| 545 |
+
"loss": 0.0276,
|
| 546 |
+
"step": 69
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.4,
|
| 550 |
+
"grad_norm": 1.0411546230316162,
|
| 551 |
+
"learning_rate": 3.421052631578947e-05,
|
| 552 |
+
"loss": 0.009,
|
| 553 |
+
"step": 70
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.4,
|
| 557 |
+
"eval_accuracy": 0.9901315789473685,
|
| 558 |
+
"eval_loss": 0.020390018820762634,
|
| 559 |
+
"eval_runtime": 7.2971,
|
| 560 |
+
"eval_samples_per_second": 41.66,
|
| 561 |
+
"eval_steps_per_second": 1.096,
|
| 562 |
+
"step": 70
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.42,
|
| 566 |
+
"grad_norm": 0.7153753638267517,
|
| 567 |
+
"learning_rate": 3.3947368421052636e-05,
|
| 568 |
+
"loss": 0.0049,
|
| 569 |
+
"step": 71
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.44,
|
| 573 |
+
"grad_norm": 4.828935623168945,
|
| 574 |
+
"learning_rate": 3.368421052631579e-05,
|
| 575 |
+
"loss": 0.0297,
|
| 576 |
+
"step": 72
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.46,
|
| 580 |
+
"grad_norm": 9.103313446044922,
|
| 581 |
+
"learning_rate": 3.342105263157895e-05,
|
| 582 |
+
"loss": 0.071,
|
| 583 |
+
"step": 73
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.48,
|
| 587 |
+
"grad_norm": 7.635372161865234,
|
| 588 |
+
"learning_rate": 3.3157894736842106e-05,
|
| 589 |
+
"loss": 0.0363,
|
| 590 |
+
"step": 74
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.5,
|
| 594 |
+
"grad_norm": 8.57288646697998,
|
| 595 |
+
"learning_rate": 3.289473684210527e-05,
|
| 596 |
+
"loss": 0.0566,
|
| 597 |
+
"step": 75
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.52,
|
| 601 |
+
"grad_norm": 8.89769172668457,
|
| 602 |
+
"learning_rate": 3.2631578947368426e-05,
|
| 603 |
+
"loss": 0.0213,
|
| 604 |
+
"step": 76
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.54,
|
| 608 |
+
"grad_norm": 2.804738998413086,
|
| 609 |
+
"learning_rate": 3.236842105263158e-05,
|
| 610 |
+
"loss": 0.0138,
|
| 611 |
+
"step": 77
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.56,
|
| 615 |
+
"grad_norm": 4.841202259063721,
|
| 616 |
+
"learning_rate": 3.210526315789474e-05,
|
| 617 |
+
"loss": 0.0181,
|
| 618 |
+
"step": 78
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.58,
|
| 622 |
+
"grad_norm": 7.109033584594727,
|
| 623 |
+
"learning_rate": 3.1842105263157895e-05,
|
| 624 |
+
"loss": 0.0332,
|
| 625 |
+
"step": 79
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.6,
|
| 629 |
+
"grad_norm": 13.540718078613281,
|
| 630 |
+
"learning_rate": 3.157894736842105e-05,
|
| 631 |
+
"loss": 0.1667,
|
| 632 |
+
"step": 80
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.6,
|
| 636 |
+
"eval_accuracy": 0.9703947368421053,
|
| 637 |
+
"eval_loss": 0.10031529515981674,
|
| 638 |
+
"eval_runtime": 6.7491,
|
| 639 |
+
"eval_samples_per_second": 45.043,
|
| 640 |
+
"eval_steps_per_second": 1.185,
|
| 641 |
+
"step": 80
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.62,
|
| 645 |
+
"grad_norm": 30.279451370239258,
|
| 646 |
+
"learning_rate": 3.131578947368421e-05,
|
| 647 |
+
"loss": 0.1139,
|
| 648 |
+
"step": 81
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.64,
|
| 652 |
+
"grad_norm": 7.229616165161133,
|
| 653 |
+
"learning_rate": 3.105263157894737e-05,
|
| 654 |
+
"loss": 0.0284,
|
| 655 |
+
"step": 82
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.66,
|
| 659 |
+
"grad_norm": 0.327736496925354,
|
| 660 |
+
"learning_rate": 3.078947368421053e-05,
|
| 661 |
+
"loss": 0.0032,
|
| 662 |
+
"step": 83
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.68,
|
| 666 |
+
"grad_norm": 0.15309709310531616,
|
| 667 |
+
"learning_rate": 3.0526315789473684e-05,
|
| 668 |
+
"loss": 0.0011,
|
| 669 |
+
"step": 84
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.7,
|
| 673 |
+
"grad_norm": 13.175533294677734,
|
| 674 |
+
"learning_rate": 3.0263157894736844e-05,
|
| 675 |
+
"loss": 0.039,
|
| 676 |
+
"step": 85
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.72,
|
| 680 |
+
"grad_norm": 24.578662872314453,
|
| 681 |
+
"learning_rate": 3e-05,
|
| 682 |
+
"loss": 0.1834,
|
| 683 |
+
"step": 86
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.74,
|
| 687 |
+
"grad_norm": 6.690978527069092,
|
| 688 |
+
"learning_rate": 2.9736842105263157e-05,
|
| 689 |
+
"loss": 0.0214,
|
| 690 |
+
"step": 87
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.76,
|
| 694 |
+
"grad_norm": 21.097963333129883,
|
| 695 |
+
"learning_rate": 2.9473684210526314e-05,
|
| 696 |
+
"loss": 0.1569,
|
| 697 |
+
"step": 88
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.78,
|
| 701 |
+
"grad_norm": 7.984706401824951,
|
| 702 |
+
"learning_rate": 2.9210526315789477e-05,
|
| 703 |
+
"loss": 0.0326,
|
| 704 |
+
"step": 89
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.8,
|
| 708 |
+
"grad_norm": 11.008134841918945,
|
| 709 |
+
"learning_rate": 2.8947368421052634e-05,
|
| 710 |
+
"loss": 0.0602,
|
| 711 |
+
"step": 90
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.8,
|
| 715 |
+
"eval_accuracy": 0.9868421052631579,
|
| 716 |
+
"eval_loss": 0.052807241678237915,
|
| 717 |
+
"eval_runtime": 7.1111,
|
| 718 |
+
"eval_samples_per_second": 42.75,
|
| 719 |
+
"eval_steps_per_second": 1.125,
|
| 720 |
+
"step": 90
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.82,
|
| 724 |
+
"grad_norm": 21.149032592773438,
|
| 725 |
+
"learning_rate": 2.868421052631579e-05,
|
| 726 |
+
"loss": 0.0265,
|
| 727 |
+
"step": 91
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.84,
|
| 731 |
+
"grad_norm": 3.4383814334869385,
|
| 732 |
+
"learning_rate": 2.842105263157895e-05,
|
| 733 |
+
"loss": 0.0059,
|
| 734 |
+
"step": 92
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.86,
|
| 738 |
+
"grad_norm": 6.291805267333984,
|
| 739 |
+
"learning_rate": 2.8157894736842106e-05,
|
| 740 |
+
"loss": 0.0177,
|
| 741 |
+
"step": 93
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 1.88,
|
| 745 |
+
"grad_norm": 1.5197041034698486,
|
| 746 |
+
"learning_rate": 2.7894736842105263e-05,
|
| 747 |
+
"loss": 0.0065,
|
| 748 |
+
"step": 94
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 1.9,
|
| 752 |
+
"grad_norm": 0.47119140625,
|
| 753 |
+
"learning_rate": 2.7631578947368426e-05,
|
| 754 |
+
"loss": 0.0018,
|
| 755 |
+
"step": 95
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.92,
|
| 759 |
+
"grad_norm": 7.311519145965576,
|
| 760 |
+
"learning_rate": 2.7368421052631583e-05,
|
| 761 |
+
"loss": 0.0235,
|
| 762 |
+
"step": 96
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.94,
|
| 766 |
+
"grad_norm": 53.3145866394043,
|
| 767 |
+
"learning_rate": 2.710526315789474e-05,
|
| 768 |
+
"loss": 0.1186,
|
| 769 |
+
"step": 97
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.96,
|
| 773 |
+
"grad_norm": 13.420350074768066,
|
| 774 |
+
"learning_rate": 2.6842105263157896e-05,
|
| 775 |
+
"loss": 0.2332,
|
| 776 |
+
"step": 98
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.98,
|
| 780 |
+
"grad_norm": 13.3897123336792,
|
| 781 |
+
"learning_rate": 2.6578947368421052e-05,
|
| 782 |
+
"loss": 0.0759,
|
| 783 |
+
"step": 99
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.0,
|
| 787 |
+
"grad_norm": 1.8416231870651245,
|
| 788 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 789 |
+
"loss": 0.0076,
|
| 790 |
+
"step": 100
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.0,
|
| 794 |
+
"eval_accuracy": 0.993421052631579,
|
| 795 |
+
"eval_loss": 0.027568014338612556,
|
| 796 |
+
"eval_runtime": 7.0992,
|
| 797 |
+
"eval_samples_per_second": 42.822,
|
| 798 |
+
"eval_steps_per_second": 1.127,
|
| 799 |
+
"step": 100
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.02,
|
| 803 |
+
"grad_norm": 6.973387241363525,
|
| 804 |
+
"learning_rate": 2.605263157894737e-05,
|
| 805 |
+
"loss": 0.0255,
|
| 806 |
+
"step": 101
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.04,
|
| 810 |
+
"grad_norm": 0.33660009503364563,
|
| 811 |
+
"learning_rate": 2.578947368421053e-05,
|
| 812 |
+
"loss": 0.0012,
|
| 813 |
+
"step": 102
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.06,
|
| 817 |
+
"grad_norm": 11.269298553466797,
|
| 818 |
+
"learning_rate": 2.5526315789473688e-05,
|
| 819 |
+
"loss": 0.1097,
|
| 820 |
+
"step": 103
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.08,
|
| 824 |
+
"grad_norm": 7.602676868438721,
|
| 825 |
+
"learning_rate": 2.5263157894736845e-05,
|
| 826 |
+
"loss": 0.034,
|
| 827 |
+
"step": 104
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 2.1,
|
| 831 |
+
"grad_norm": 7.900285720825195,
|
| 832 |
+
"learning_rate": 2.5e-05,
|
| 833 |
+
"loss": 0.0399,
|
| 834 |
+
"step": 105
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.12,
|
| 838 |
+
"grad_norm": 4.07220983505249,
|
| 839 |
+
"learning_rate": 2.4736842105263158e-05,
|
| 840 |
+
"loss": 0.0055,
|
| 841 |
+
"step": 106
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.14,
|
| 845 |
+
"grad_norm": 0.09085848927497864,
|
| 846 |
+
"learning_rate": 2.4473684210526318e-05,
|
| 847 |
+
"loss": 0.0005,
|
| 848 |
+
"step": 107
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.16,
|
| 852 |
+
"grad_norm": 0.5666660666465759,
|
| 853 |
+
"learning_rate": 2.4210526315789474e-05,
|
| 854 |
+
"loss": 0.0029,
|
| 855 |
+
"step": 108
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 2.18,
|
| 859 |
+
"grad_norm": 12.90062427520752,
|
| 860 |
+
"learning_rate": 2.394736842105263e-05,
|
| 861 |
+
"loss": 0.051,
|
| 862 |
+
"step": 109
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 2.2,
|
| 866 |
+
"grad_norm": 0.24952377378940582,
|
| 867 |
+
"learning_rate": 2.368421052631579e-05,
|
| 868 |
+
"loss": 0.0007,
|
| 869 |
+
"step": 110
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 2.2,
|
| 873 |
+
"eval_accuracy": 0.9835526315789473,
|
| 874 |
+
"eval_loss": 0.05723079293966293,
|
| 875 |
+
"eval_runtime": 7.2757,
|
| 876 |
+
"eval_samples_per_second": 41.783,
|
| 877 |
+
"eval_steps_per_second": 1.1,
|
| 878 |
+
"step": 110
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.22,
|
| 882 |
+
"grad_norm": 2.654731512069702,
|
| 883 |
+
"learning_rate": 2.342105263157895e-05,
|
| 884 |
+
"loss": 0.0036,
|
| 885 |
+
"step": 111
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.24,
|
| 889 |
+
"grad_norm": 0.5114130973815918,
|
| 890 |
+
"learning_rate": 2.3157894736842107e-05,
|
| 891 |
+
"loss": 0.0012,
|
| 892 |
+
"step": 112
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 2.26,
|
| 896 |
+
"grad_norm": 9.55749797821045,
|
| 897 |
+
"learning_rate": 2.2894736842105263e-05,
|
| 898 |
+
"loss": 0.0195,
|
| 899 |
+
"step": 113
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.28,
|
| 903 |
+
"grad_norm": 11.102432250976562,
|
| 904 |
+
"learning_rate": 2.2631578947368423e-05,
|
| 905 |
+
"loss": 0.0204,
|
| 906 |
+
"step": 114
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 2.3,
|
| 910 |
+
"grad_norm": 2.684544324874878,
|
| 911 |
+
"learning_rate": 2.236842105263158e-05,
|
| 912 |
+
"loss": 0.0047,
|
| 913 |
+
"step": 115
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.32,
|
| 917 |
+
"grad_norm": 0.22236469388008118,
|
| 918 |
+
"learning_rate": 2.2105263157894736e-05,
|
| 919 |
+
"loss": 0.0008,
|
| 920 |
+
"step": 116
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 2.34,
|
| 924 |
+
"grad_norm": 8.926434516906738,
|
| 925 |
+
"learning_rate": 2.1842105263157896e-05,
|
| 926 |
+
"loss": 0.0223,
|
| 927 |
+
"step": 117
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.36,
|
| 931 |
+
"grad_norm": 3.3315141201019287,
|
| 932 |
+
"learning_rate": 2.1578947368421053e-05,
|
| 933 |
+
"loss": 0.0117,
|
| 934 |
+
"step": 118
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 2.38,
|
| 938 |
+
"grad_norm": 10.430353164672852,
|
| 939 |
+
"learning_rate": 2.1315789473684212e-05,
|
| 940 |
+
"loss": 0.0303,
|
| 941 |
+
"step": 119
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 2.4,
|
| 945 |
+
"grad_norm": 0.15297500789165497,
|
| 946 |
+
"learning_rate": 2.105263157894737e-05,
|
| 947 |
+
"loss": 0.0008,
|
| 948 |
+
"step": 120
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 2.4,
|
| 952 |
+
"eval_accuracy": 0.9967105263157895,
|
| 953 |
+
"eval_loss": 0.009093067608773708,
|
| 954 |
+
"eval_runtime": 6.8131,
|
| 955 |
+
"eval_samples_per_second": 44.62,
|
| 956 |
+
"eval_steps_per_second": 1.174,
|
| 957 |
+
"step": 120
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.42,
|
| 961 |
+
"grad_norm": 0.6949332356452942,
|
| 962 |
+
"learning_rate": 2.078947368421053e-05,
|
| 963 |
+
"loss": 0.0018,
|
| 964 |
+
"step": 121
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 2.44,
|
| 968 |
+
"grad_norm": 0.10584941506385803,
|
| 969 |
+
"learning_rate": 2.0526315789473685e-05,
|
| 970 |
+
"loss": 0.0006,
|
| 971 |
+
"step": 122
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.46,
|
| 975 |
+
"grad_norm": 2.782750368118286,
|
| 976 |
+
"learning_rate": 2.0263157894736842e-05,
|
| 977 |
+
"loss": 0.0064,
|
| 978 |
+
"step": 123
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 2.48,
|
| 982 |
+
"grad_norm": 0.1843283325433731,
|
| 983 |
+
"learning_rate": 2e-05,
|
| 984 |
+
"loss": 0.0007,
|
| 985 |
+
"step": 124
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 2.5,
|
| 989 |
+
"grad_norm": 19.077423095703125,
|
| 990 |
+
"learning_rate": 1.9736842105263158e-05,
|
| 991 |
+
"loss": 0.1095,
|
| 992 |
+
"step": 125
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.52,
|
| 996 |
+
"grad_norm": 9.631857872009277,
|
| 997 |
+
"learning_rate": 1.9473684210526315e-05,
|
| 998 |
+
"loss": 0.0286,
|
| 999 |
+
"step": 126
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.54,
|
| 1003 |
+
"grad_norm": 2.46711802482605,
|
| 1004 |
+
"learning_rate": 1.9210526315789474e-05,
|
| 1005 |
+
"loss": 0.0042,
|
| 1006 |
+
"step": 127
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 2.56,
|
| 1010 |
+
"grad_norm": 0.9471967220306396,
|
| 1011 |
+
"learning_rate": 1.8947368421052634e-05,
|
| 1012 |
+
"loss": 0.0038,
|
| 1013 |
+
"step": 128
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 2.58,
|
| 1017 |
+
"grad_norm": 0.1509537547826767,
|
| 1018 |
+
"learning_rate": 1.868421052631579e-05,
|
| 1019 |
+
"loss": 0.0009,
|
| 1020 |
+
"step": 129
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 2.6,
|
| 1024 |
+
"grad_norm": 16.95519256591797,
|
| 1025 |
+
"learning_rate": 1.8421052631578947e-05,
|
| 1026 |
+
"loss": 0.1277,
|
| 1027 |
+
"step": 130
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 2.6,
|
| 1031 |
+
"eval_accuracy": 0.9967105263157895,
|
| 1032 |
+
"eval_loss": 0.004848334938287735,
|
| 1033 |
+
"eval_runtime": 7.1393,
|
| 1034 |
+
"eval_samples_per_second": 42.581,
|
| 1035 |
+
"eval_steps_per_second": 1.121,
|
| 1036 |
+
"step": 130
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 2.62,
|
| 1040 |
+
"grad_norm": 0.3852033317089081,
|
| 1041 |
+
"learning_rate": 1.8157894736842107e-05,
|
| 1042 |
+
"loss": 0.0009,
|
| 1043 |
+
"step": 131
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.64,
|
| 1047 |
+
"grad_norm": 0.23005267977714539,
|
| 1048 |
+
"learning_rate": 1.7894736842105264e-05,
|
| 1049 |
+
"loss": 0.0006,
|
| 1050 |
+
"step": 132
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 2.66,
|
| 1054 |
+
"grad_norm": 2.9750654697418213,
|
| 1055 |
+
"learning_rate": 1.763157894736842e-05,
|
| 1056 |
+
"loss": 0.0096,
|
| 1057 |
+
"step": 133
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 2.68,
|
| 1061 |
+
"grad_norm": 6.330101490020752,
|
| 1062 |
+
"learning_rate": 1.736842105263158e-05,
|
| 1063 |
+
"loss": 0.0224,
|
| 1064 |
+
"step": 134
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.7,
|
| 1068 |
+
"grad_norm": 0.2537049949169159,
|
| 1069 |
+
"learning_rate": 1.7105263157894737e-05,
|
| 1070 |
+
"loss": 0.001,
|
| 1071 |
+
"step": 135
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 2.72,
|
| 1075 |
+
"grad_norm": 0.49432969093322754,
|
| 1076 |
+
"learning_rate": 1.6842105263157896e-05,
|
| 1077 |
+
"loss": 0.002,
|
| 1078 |
+
"step": 136
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 2.74,
|
| 1082 |
+
"grad_norm": 0.878919243812561,
|
| 1083 |
+
"learning_rate": 1.6578947368421053e-05,
|
| 1084 |
+
"loss": 0.0022,
|
| 1085 |
+
"step": 137
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 2.76,
|
| 1089 |
+
"grad_norm": 0.4538193941116333,
|
| 1090 |
+
"learning_rate": 1.6315789473684213e-05,
|
| 1091 |
+
"loss": 0.0016,
|
| 1092 |
+
"step": 138
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 2.78,
|
| 1096 |
+
"grad_norm": 13.346075057983398,
|
| 1097 |
+
"learning_rate": 1.605263157894737e-05,
|
| 1098 |
+
"loss": 0.0384,
|
| 1099 |
+
"step": 139
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 2.8,
|
| 1103 |
+
"grad_norm": 0.42253851890563965,
|
| 1104 |
+
"learning_rate": 1.5789473684210526e-05,
|
| 1105 |
+
"loss": 0.0007,
|
| 1106 |
+
"step": 140
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 2.8,
|
| 1110 |
+
"eval_accuracy": 0.9736842105263158,
|
| 1111 |
+
"eval_loss": 0.08703595399856567,
|
| 1112 |
+
"eval_runtime": 7.1737,
|
| 1113 |
+
"eval_samples_per_second": 42.377,
|
| 1114 |
+
"eval_steps_per_second": 1.115,
|
| 1115 |
+
"step": 140
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.82,
|
| 1119 |
+
"grad_norm": 8.999384880065918,
|
| 1120 |
+
"learning_rate": 1.5526315789473686e-05,
|
| 1121 |
+
"loss": 0.041,
|
| 1122 |
+
"step": 141
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.84,
|
| 1126 |
+
"grad_norm": 0.34943702816963196,
|
| 1127 |
+
"learning_rate": 1.5263157894736842e-05,
|
| 1128 |
+
"loss": 0.0008,
|
| 1129 |
+
"step": 142
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.86,
|
| 1133 |
+
"grad_norm": 41.58428192138672,
|
| 1134 |
+
"learning_rate": 1.5e-05,
|
| 1135 |
+
"loss": 0.0352,
|
| 1136 |
+
"step": 143
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.88,
|
| 1140 |
+
"grad_norm": 1.358328104019165,
|
| 1141 |
+
"learning_rate": 1.4736842105263157e-05,
|
| 1142 |
+
"loss": 0.004,
|
| 1143 |
+
"step": 144
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.9,
|
| 1147 |
+
"grad_norm": 26.084732055664062,
|
| 1148 |
+
"learning_rate": 1.4473684210526317e-05,
|
| 1149 |
+
"loss": 0.1456,
|
| 1150 |
+
"step": 145
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.92,
|
| 1154 |
+
"grad_norm": 4.426596164703369,
|
| 1155 |
+
"learning_rate": 1.4210526315789475e-05,
|
| 1156 |
+
"loss": 0.0056,
|
| 1157 |
+
"step": 146
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.94,
|
| 1161 |
+
"grad_norm": 7.729344367980957,
|
| 1162 |
+
"learning_rate": 1.3947368421052631e-05,
|
| 1163 |
+
"loss": 0.0536,
|
| 1164 |
+
"step": 147
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.96,
|
| 1168 |
+
"grad_norm": 0.49353259801864624,
|
| 1169 |
+
"learning_rate": 1.3684210526315791e-05,
|
| 1170 |
+
"loss": 0.0035,
|
| 1171 |
+
"step": 148
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.98,
|
| 1175 |
+
"grad_norm": 15.301690101623535,
|
| 1176 |
+
"learning_rate": 1.3421052631578948e-05,
|
| 1177 |
+
"loss": 0.0138,
|
| 1178 |
+
"step": 149
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.0,
|
| 1182 |
+
"grad_norm": 0.520106852054596,
|
| 1183 |
+
"learning_rate": 1.3157894736842106e-05,
|
| 1184 |
+
"loss": 0.0008,
|
| 1185 |
+
"step": 150
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.0,
|
| 1189 |
+
"eval_accuracy": 0.993421052631579,
|
| 1190 |
+
"eval_loss": 0.005907059647142887,
|
| 1191 |
+
"eval_runtime": 7.2044,
|
| 1192 |
+
"eval_samples_per_second": 42.197,
|
| 1193 |
+
"eval_steps_per_second": 1.11,
|
| 1194 |
+
"step": 150
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 3.02,
|
| 1198 |
+
"grad_norm": 0.7136516571044922,
|
| 1199 |
+
"learning_rate": 1.2894736842105264e-05,
|
| 1200 |
+
"loss": 0.0028,
|
| 1201 |
+
"step": 151
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 3.04,
|
| 1205 |
+
"grad_norm": 0.28995925188064575,
|
| 1206 |
+
"learning_rate": 1.2631578947368422e-05,
|
| 1207 |
+
"loss": 0.0006,
|
| 1208 |
+
"step": 152
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 3.06,
|
| 1212 |
+
"grad_norm": 9.869847297668457,
|
| 1213 |
+
"learning_rate": 1.2368421052631579e-05,
|
| 1214 |
+
"loss": 0.0342,
|
| 1215 |
+
"step": 153
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 3.08,
|
| 1219 |
+
"grad_norm": 0.20523004233837128,
|
| 1220 |
+
"learning_rate": 1.2105263157894737e-05,
|
| 1221 |
+
"loss": 0.0005,
|
| 1222 |
+
"step": 154
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 3.1,
|
| 1226 |
+
"grad_norm": 8.038060188293457,
|
| 1227 |
+
"learning_rate": 1.1842105263157895e-05,
|
| 1228 |
+
"loss": 0.0178,
|
| 1229 |
+
"step": 155
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 3.12,
|
| 1233 |
+
"grad_norm": 1.2011505365371704,
|
| 1234 |
+
"learning_rate": 1.1578947368421053e-05,
|
| 1235 |
+
"loss": 0.002,
|
| 1236 |
+
"step": 156
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 3.14,
|
| 1240 |
+
"grad_norm": 1.0488544702529907,
|
| 1241 |
+
"learning_rate": 1.1315789473684212e-05,
|
| 1242 |
+
"loss": 0.0025,
|
| 1243 |
+
"step": 157
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 3.16,
|
| 1247 |
+
"grad_norm": 0.24755439162254333,
|
| 1248 |
+
"learning_rate": 1.1052631578947368e-05,
|
| 1249 |
+
"loss": 0.0005,
|
| 1250 |
+
"step": 158
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 3.18,
|
| 1254 |
+
"grad_norm": 0.37720346450805664,
|
| 1255 |
+
"learning_rate": 1.0789473684210526e-05,
|
| 1256 |
+
"loss": 0.0007,
|
| 1257 |
+
"step": 159
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 3.2,
|
| 1261 |
+
"grad_norm": 0.054420772939920425,
|
| 1262 |
+
"learning_rate": 1.0526315789473684e-05,
|
| 1263 |
+
"loss": 0.0003,
|
| 1264 |
+
"step": 160
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 3.2,
|
| 1268 |
+
"eval_accuracy": 0.9967105263157895,
|
| 1269 |
+
"eval_loss": 0.012113516218960285,
|
| 1270 |
+
"eval_runtime": 7.4779,
|
| 1271 |
+
"eval_samples_per_second": 40.653,
|
| 1272 |
+
"eval_steps_per_second": 1.07,
|
| 1273 |
+
"step": 160
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 3.22,
|
| 1277 |
+
"grad_norm": 3.170163631439209,
|
| 1278 |
+
"learning_rate": 1.0263157894736843e-05,
|
| 1279 |
+
"loss": 0.0048,
|
| 1280 |
+
"step": 161
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 3.24,
|
| 1284 |
+
"grad_norm": 9.659165382385254,
|
| 1285 |
+
"learning_rate": 1e-05,
|
| 1286 |
+
"loss": 0.006,
|
| 1287 |
+
"step": 162
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 3.26,
|
| 1291 |
+
"grad_norm": 0.07884739339351654,
|
| 1292 |
+
"learning_rate": 9.736842105263157e-06,
|
| 1293 |
+
"loss": 0.0003,
|
| 1294 |
+
"step": 163
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 3.28,
|
| 1298 |
+
"grad_norm": 1.1162643432617188,
|
| 1299 |
+
"learning_rate": 9.473684210526317e-06,
|
| 1300 |
+
"loss": 0.0019,
|
| 1301 |
+
"step": 164
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 3.3,
|
| 1305 |
+
"grad_norm": 0.3206580877304077,
|
| 1306 |
+
"learning_rate": 9.210526315789474e-06,
|
| 1307 |
+
"loss": 0.001,
|
| 1308 |
+
"step": 165
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 3.32,
|
| 1312 |
+
"grad_norm": 15.26516056060791,
|
| 1313 |
+
"learning_rate": 8.947368421052632e-06,
|
| 1314 |
+
"loss": 0.0467,
|
| 1315 |
+
"step": 166
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 3.34,
|
| 1319 |
+
"grad_norm": 0.6477659940719604,
|
| 1320 |
+
"learning_rate": 8.68421052631579e-06,
|
| 1321 |
+
"loss": 0.002,
|
| 1322 |
+
"step": 167
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 3.36,
|
| 1326 |
+
"grad_norm": 16.165756225585938,
|
| 1327 |
+
"learning_rate": 8.421052631578948e-06,
|
| 1328 |
+
"loss": 0.0486,
|
| 1329 |
+
"step": 168
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 3.38,
|
| 1333 |
+
"grad_norm": 0.3942013382911682,
|
| 1334 |
+
"learning_rate": 8.157894736842106e-06,
|
| 1335 |
+
"loss": 0.001,
|
| 1336 |
+
"step": 169
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 3.4,
|
| 1340 |
+
"grad_norm": 0.2746890187263489,
|
| 1341 |
+
"learning_rate": 7.894736842105263e-06,
|
| 1342 |
+
"loss": 0.0008,
|
| 1343 |
+
"step": 170
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 3.4,
|
| 1347 |
+
"eval_accuracy": 0.993421052631579,
|
| 1348 |
+
"eval_loss": 0.02530418336391449,
|
| 1349 |
+
"eval_runtime": 6.8199,
|
| 1350 |
+
"eval_samples_per_second": 44.575,
|
| 1351 |
+
"eval_steps_per_second": 1.173,
|
| 1352 |
+
"step": 170
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 3.42,
|
| 1356 |
+
"grad_norm": 0.1466667354106903,
|
| 1357 |
+
"learning_rate": 7.631578947368421e-06,
|
| 1358 |
+
"loss": 0.0006,
|
| 1359 |
+
"step": 171
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 3.44,
|
| 1363 |
+
"grad_norm": 0.35398414731025696,
|
| 1364 |
+
"learning_rate": 7.3684210526315784e-06,
|
| 1365 |
+
"loss": 0.0008,
|
| 1366 |
+
"step": 172
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 3.46,
|
| 1370 |
+
"grad_norm": 0.03358185663819313,
|
| 1371 |
+
"learning_rate": 7.1052631578947375e-06,
|
| 1372 |
+
"loss": 0.0001,
|
| 1373 |
+
"step": 173
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 3.48,
|
| 1377 |
+
"grad_norm": 0.04026187211275101,
|
| 1378 |
+
"learning_rate": 6.842105263157896e-06,
|
| 1379 |
+
"loss": 0.0002,
|
| 1380 |
+
"step": 174
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 3.5,
|
| 1384 |
+
"grad_norm": 0.10377287119626999,
|
| 1385 |
+
"learning_rate": 6.578947368421053e-06,
|
| 1386 |
+
"loss": 0.0008,
|
| 1387 |
+
"step": 175
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 3.52,
|
| 1391 |
+
"grad_norm": 3.1232450008392334,
|
| 1392 |
+
"learning_rate": 6.315789473684211e-06,
|
| 1393 |
+
"loss": 0.0062,
|
| 1394 |
+
"step": 176
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 3.54,
|
| 1398 |
+
"grad_norm": 0.0659070760011673,
|
| 1399 |
+
"learning_rate": 6.0526315789473685e-06,
|
| 1400 |
+
"loss": 0.0002,
|
| 1401 |
+
"step": 177
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 3.56,
|
| 1405 |
+
"grad_norm": 0.4386169910430908,
|
| 1406 |
+
"learning_rate": 5.789473684210527e-06,
|
| 1407 |
+
"loss": 0.0007,
|
| 1408 |
+
"step": 178
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 3.58,
|
| 1412 |
+
"grad_norm": 0.09403068572282791,
|
| 1413 |
+
"learning_rate": 5.526315789473684e-06,
|
| 1414 |
+
"loss": 0.0001,
|
| 1415 |
+
"step": 179
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 3.6,
|
| 1419 |
+
"grad_norm": 0.011226998642086983,
|
| 1420 |
+
"learning_rate": 5.263157894736842e-06,
|
| 1421 |
+
"loss": 0.0,
|
| 1422 |
+
"step": 180
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 3.6,
|
| 1426 |
+
"eval_accuracy": 0.9967105263157895,
|
| 1427 |
+
"eval_loss": 0.012991190887987614,
|
| 1428 |
+
"eval_runtime": 7.2339,
|
| 1429 |
+
"eval_samples_per_second": 42.024,
|
| 1430 |
+
"eval_steps_per_second": 1.106,
|
| 1431 |
+
"step": 180
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 3.62,
|
| 1435 |
+
"grad_norm": 0.3960583209991455,
|
| 1436 |
+
"learning_rate": 5e-06,
|
| 1437 |
+
"loss": 0.0013,
|
| 1438 |
+
"step": 181
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 3.64,
|
| 1442 |
+
"grad_norm": 0.028123315423727036,
|
| 1443 |
+
"learning_rate": 4.736842105263159e-06,
|
| 1444 |
+
"loss": 0.0001,
|
| 1445 |
+
"step": 182
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 3.66,
|
| 1449 |
+
"grad_norm": 3.7100679874420166,
|
| 1450 |
+
"learning_rate": 4.473684210526316e-06,
|
| 1451 |
+
"loss": 0.0062,
|
| 1452 |
+
"step": 183
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 3.68,
|
| 1456 |
+
"grad_norm": 0.027926042675971985,
|
| 1457 |
+
"learning_rate": 4.210526315789474e-06,
|
| 1458 |
+
"loss": 0.0001,
|
| 1459 |
+
"step": 184
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 3.7,
|
| 1463 |
+
"grad_norm": 1.727767825126648,
|
| 1464 |
+
"learning_rate": 3.9473684210526315e-06,
|
| 1465 |
+
"loss": 0.0022,
|
| 1466 |
+
"step": 185
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 3.72,
|
| 1470 |
+
"grad_norm": 1.4793553352355957,
|
| 1471 |
+
"learning_rate": 3.6842105263157892e-06,
|
| 1472 |
+
"loss": 0.0011,
|
| 1473 |
+
"step": 186
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 3.74,
|
| 1477 |
+
"grad_norm": 2.847693681716919,
|
| 1478 |
+
"learning_rate": 3.421052631578948e-06,
|
| 1479 |
+
"loss": 0.004,
|
| 1480 |
+
"step": 187
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 3.76,
|
| 1484 |
+
"grad_norm": 0.017291821539402008,
|
| 1485 |
+
"learning_rate": 3.1578947368421056e-06,
|
| 1486 |
+
"loss": 0.0001,
|
| 1487 |
+
"step": 188
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 3.78,
|
| 1491 |
+
"grad_norm": 0.6957990527153015,
|
| 1492 |
+
"learning_rate": 2.8947368421052634e-06,
|
| 1493 |
+
"loss": 0.001,
|
| 1494 |
+
"step": 189
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.8,
|
| 1498 |
+
"grad_norm": 0.032589249312877655,
|
| 1499 |
+
"learning_rate": 2.631578947368421e-06,
|
| 1500 |
+
"loss": 0.0001,
|
| 1501 |
+
"step": 190
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 3.8,
|
| 1505 |
+
"eval_accuracy": 0.993421052631579,
|
| 1506 |
+
"eval_loss": 0.009236541576683521,
|
| 1507 |
+
"eval_runtime": 6.8219,
|
| 1508 |
+
"eval_samples_per_second": 44.563,
|
| 1509 |
+
"eval_steps_per_second": 1.173,
|
| 1510 |
+
"step": 190
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 3.82,
|
| 1514 |
+
"grad_norm": 0.5736590027809143,
|
| 1515 |
+
"learning_rate": 2.3684210526315793e-06,
|
| 1516 |
+
"loss": 0.0013,
|
| 1517 |
+
"step": 191
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 3.84,
|
| 1521 |
+
"grad_norm": 0.4599183201789856,
|
| 1522 |
+
"learning_rate": 2.105263157894737e-06,
|
| 1523 |
+
"loss": 0.0016,
|
| 1524 |
+
"step": 192
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 3.86,
|
| 1528 |
+
"grad_norm": 0.0121466601267457,
|
| 1529 |
+
"learning_rate": 1.8421052631578946e-06,
|
| 1530 |
+
"loss": 0.0,
|
| 1531 |
+
"step": 193
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 3.88,
|
| 1535 |
+
"grad_norm": 2.155550479888916,
|
| 1536 |
+
"learning_rate": 1.5789473684210528e-06,
|
| 1537 |
+
"loss": 0.0048,
|
| 1538 |
+
"step": 194
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 3.9,
|
| 1542 |
+
"grad_norm": 0.3076814115047455,
|
| 1543 |
+
"learning_rate": 1.3157894736842106e-06,
|
| 1544 |
+
"loss": 0.0006,
|
| 1545 |
+
"step": 195
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 3.92,
|
| 1549 |
+
"grad_norm": 0.05223652720451355,
|
| 1550 |
+
"learning_rate": 1.0526315789473685e-06,
|
| 1551 |
+
"loss": 0.0002,
|
| 1552 |
+
"step": 196
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 3.94,
|
| 1556 |
+
"grad_norm": 0.10159199684858322,
|
| 1557 |
+
"learning_rate": 7.894736842105264e-07,
|
| 1558 |
+
"loss": 0.0001,
|
| 1559 |
+
"step": 197
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 3.96,
|
| 1563 |
+
"grad_norm": 7.128465175628662,
|
| 1564 |
+
"learning_rate": 5.263157894736843e-07,
|
| 1565 |
+
"loss": 0.0243,
|
| 1566 |
+
"step": 198
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 3.98,
|
| 1570 |
+
"grad_norm": 0.21553558111190796,
|
| 1571 |
+
"learning_rate": 2.6315789473684213e-07,
|
| 1572 |
+
"loss": 0.0007,
|
| 1573 |
+
"step": 199
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 4.0,
|
| 1577 |
+
"grad_norm": 0.11059165745973587,
|
| 1578 |
+
"learning_rate": 0.0,
|
| 1579 |
+
"loss": 0.0002,
|
| 1580 |
+
"step": 200
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 4.0,
|
| 1584 |
+
"eval_accuracy": 0.9967105263157895,
|
| 1585 |
+
"eval_loss": 0.011121152900159359,
|
| 1586 |
+
"eval_runtime": 7.2058,
|
| 1587 |
+
"eval_samples_per_second": 42.188,
|
| 1588 |
+
"eval_steps_per_second": 1.11,
|
| 1589 |
+
"step": 200
|
| 1590 |
+
}
|
| 1591 |
+
],
|
| 1592 |
+
"logging_steps": 1,
|
| 1593 |
+
"max_steps": 200,
|
| 1594 |
+
"num_input_tokens_seen": 0,
|
| 1595 |
+
"num_train_epochs": 4,
|
| 1596 |
+
"save_steps": 10,
|
| 1597 |
+
"total_flos": 6.201834183515013e+17,
|
| 1598 |
+
"train_batch_size": 40,
|
| 1599 |
+
"trial_name": null,
|
| 1600 |
+
"trial_params": null
|
| 1601 |
+
}
|
checkpoint-200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d12326b817ccf86422554696b6e15d756b4174e67e313ecfbe14aadf3bb73244
|
| 3 |
+
size 4856
|
config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Organika/sdxl-detector",
|
| 3 |
+
"_num_labels": 2,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"SwinForImageClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.0,
|
| 8 |
+
"depths": [
|
| 9 |
+
2,
|
| 10 |
+
2,
|
| 11 |
+
18,
|
| 12 |
+
2
|
| 13 |
+
],
|
| 14 |
+
"drop_path_rate": 0.1,
|
| 15 |
+
"embed_dim": 128,
|
| 16 |
+
"encoder_stride": 32,
|
| 17 |
+
"hidden_act": "gelu",
|
| 18 |
+
"hidden_dropout_prob": 0.0,
|
| 19 |
+
"hidden_size": 1024,
|
| 20 |
+
"id2label": {
|
| 21 |
+
"0": "artificial",
|
| 22 |
+
"1": "human"
|
| 23 |
+
},
|
| 24 |
+
"image_size": 224,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"label2id": {
|
| 27 |
+
"artificial": 0,
|
| 28 |
+
"human": 1
|
| 29 |
+
},
|
| 30 |
+
"layer_norm_eps": 1e-05,
|
| 31 |
+
"max_length": 128,
|
| 32 |
+
"mlp_ratio": 4.0,
|
| 33 |
+
"model_type": "swin",
|
| 34 |
+
"num_channels": 3,
|
| 35 |
+
"num_heads": [
|
| 36 |
+
4,
|
| 37 |
+
8,
|
| 38 |
+
16,
|
| 39 |
+
32
|
| 40 |
+
],
|
| 41 |
+
"num_layers": 4,
|
| 42 |
+
"out_features": [
|
| 43 |
+
"stage4"
|
| 44 |
+
],
|
| 45 |
+
"out_indices": [
|
| 46 |
+
4
|
| 47 |
+
],
|
| 48 |
+
"padding": "max_length",
|
| 49 |
+
"patch_size": 4,
|
| 50 |
+
"path_norm": true,
|
| 51 |
+
"problem_type": "single_label_classification",
|
| 52 |
+
"qkv_bias": true,
|
| 53 |
+
"stage_names": [
|
| 54 |
+
"stem",
|
| 55 |
+
"stage1",
|
| 56 |
+
"stage2",
|
| 57 |
+
"stage3",
|
| 58 |
+
"stage4"
|
| 59 |
+
],
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.39.3",
|
| 62 |
+
"use_absolute_embeddings": false,
|
| 63 |
+
"window_size": 7
|
| 64 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86a9c215382ea7372ff3d6cc70c3c6268b24b661c627fbfe8ef12ac3abd2a13a
|
| 3 |
+
size 347498816
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.485,
|
| 21 |
+
0.456,
|
| 22 |
+
0.406
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.229,
|
| 27 |
+
0.224,
|
| 28 |
+
0.225
|
| 29 |
+
],
|
| 30 |
+
"resample": 3,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d12326b817ccf86422554696b6e15d756b4174e67e313ecfbe14aadf3bb73244
|
| 3 |
+
size 4856
|