add tokenizer
Browse files- checkpoint-1000/config.json +109 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/preprocessor_config.json +9 -0
- checkpoint-1000/pytorch_model.bin +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scaler.pt +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/trainer_state.json +1234 -0
- checkpoint-1000/training_args.bin +3 -0
- checkpoint-500/config.json +109 -0
- checkpoint-500/optimizer.pt +3 -0
- checkpoint-500/preprocessor_config.json +9 -0
- checkpoint-500/pytorch_model.bin +3 -0
- checkpoint-500/rng_state.pth +3 -0
- checkpoint-500/scaler.pt +3 -0
- checkpoint-500/scheduler.pt +3 -0
- checkpoint-500/trainer_state.json +625 -0
- checkpoint-500/training_args.bin +3 -0
checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-base-it-voxpopuli",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForCTC"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.1,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 256,
|
| 15 |
+
"contrastive_logits_temperature": 0.1,
|
| 16 |
+
"conv_bias": false,
|
| 17 |
+
"conv_dim": [
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512
|
| 25 |
+
],
|
| 26 |
+
"conv_kernel": [
|
| 27 |
+
10,
|
| 28 |
+
3,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
2,
|
| 33 |
+
2
|
| 34 |
+
],
|
| 35 |
+
"conv_stride": [
|
| 36 |
+
5,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"ctc_loss_reduction": "mean",
|
| 45 |
+
"ctc_zero_infinity": false,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": false,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "group",
|
| 52 |
+
"feat_proj_dropout": 0.1,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.1,
|
| 55 |
+
"gradient_checkpointing": false,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.1,
|
| 58 |
+
"hidden_dropout_prob": 0.1,
|
| 59 |
+
"hidden_size": 768,
|
| 60 |
+
"initializer_range": 0.02,
|
| 61 |
+
"intermediate_size": 3072,
|
| 62 |
+
"layer_norm_eps": 1e-05,
|
| 63 |
+
"layerdrop": 0.1,
|
| 64 |
+
"mask_feature_length": 10,
|
| 65 |
+
"mask_feature_min_masks": 0,
|
| 66 |
+
"mask_feature_prob": 0.0,
|
| 67 |
+
"mask_time_length": 10,
|
| 68 |
+
"mask_time_min_masks": 2,
|
| 69 |
+
"mask_time_prob": 0.05,
|
| 70 |
+
"model_type": "wav2vec2",
|
| 71 |
+
"num_adapter_layers": 3,
|
| 72 |
+
"num_attention_heads": 12,
|
| 73 |
+
"num_codevector_groups": 2,
|
| 74 |
+
"num_codevectors_per_group": 320,
|
| 75 |
+
"num_conv_pos_embedding_groups": 16,
|
| 76 |
+
"num_conv_pos_embeddings": 128,
|
| 77 |
+
"num_feat_extract_layers": 7,
|
| 78 |
+
"num_hidden_layers": 12,
|
| 79 |
+
"num_negatives": 100,
|
| 80 |
+
"output_hidden_size": 768,
|
| 81 |
+
"pad_token_id": 25,
|
| 82 |
+
"proj_codevector_dim": 256,
|
| 83 |
+
"tdnn_dilation": [
|
| 84 |
+
1,
|
| 85 |
+
2,
|
| 86 |
+
3,
|
| 87 |
+
1,
|
| 88 |
+
1
|
| 89 |
+
],
|
| 90 |
+
"tdnn_dim": [
|
| 91 |
+
512,
|
| 92 |
+
512,
|
| 93 |
+
512,
|
| 94 |
+
512,
|
| 95 |
+
1500
|
| 96 |
+
],
|
| 97 |
+
"tdnn_kernel": [
|
| 98 |
+
5,
|
| 99 |
+
3,
|
| 100 |
+
3,
|
| 101 |
+
1,
|
| 102 |
+
1
|
| 103 |
+
],
|
| 104 |
+
"torch_dtype": "float32",
|
| 105 |
+
"transformers_version": "4.16.1",
|
| 106 |
+
"use_weighted_layer_sum": false,
|
| 107 |
+
"vocab_size": 32,
|
| 108 |
+
"xvector_output_dim": 512
|
| 109 |
+
}
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:756d616dea6333b020fcf966b9e165b838b91da46642986c3558f9a864af2e52
|
| 3 |
+
size 721685265
|
checkpoint-1000/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": false,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
checkpoint-1000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10c2b84917b1857d55f5a4d2225cefc3ffa1cf416c8b2b1f2b19bb5b6706209e
|
| 3 |
+
size 377670039
|
checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b087b4fc4ab07baf9801f2880dccc210ef12179b734d6c190e30af73f5c44b1c
|
| 3 |
+
size 14567
|
checkpoint-1000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:020228eb3ae095194ff32686b7e36027b299ba9209b4ba69c90b4ac3aaf606cd
|
| 3 |
+
size 559
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42571f4e0abc1ba5c0353b7a97b263ea72c970799a16a8b489420ffa80a0f6b7
|
| 3 |
+
size 623
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,1234 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.18625442354255914,
|
| 5 |
+
"global_step": 1000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 1e-08,
|
| 13 |
+
"loss": 24.0683,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.0,
|
| 18 |
+
"learning_rate": 3.5e-08,
|
| 19 |
+
"loss": 17.9067,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.0,
|
| 24 |
+
"learning_rate": 6.000000000000001e-08,
|
| 25 |
+
"loss": 16.9004,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.0,
|
| 30 |
+
"learning_rate": 8.500000000000001e-08,
|
| 31 |
+
"loss": 16.8411,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.0,
|
| 36 |
+
"learning_rate": 1.1e-07,
|
| 37 |
+
"loss": 15.0697,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.01,
|
| 42 |
+
"learning_rate": 1.35e-07,
|
| 43 |
+
"loss": 15.2061,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01,
|
| 48 |
+
"learning_rate": 1.6e-07,
|
| 49 |
+
"loss": 14.8504,
|
| 50 |
+
"step": 35
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.01,
|
| 54 |
+
"learning_rate": 1.85e-07,
|
| 55 |
+
"loss": 14.6816,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.01,
|
| 60 |
+
"learning_rate": 2.1000000000000003e-07,
|
| 61 |
+
"loss": 14.9665,
|
| 62 |
+
"step": 45
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.01,
|
| 66 |
+
"learning_rate": 2.3500000000000003e-07,
|
| 67 |
+
"loss": 14.3427,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.01,
|
| 72 |
+
"learning_rate": 2.6e-07,
|
| 73 |
+
"loss": 18.4574,
|
| 74 |
+
"step": 55
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.01,
|
| 78 |
+
"learning_rate": 2.8e-07,
|
| 79 |
+
"loss": 18.379,
|
| 80 |
+
"step": 60
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.01,
|
| 84 |
+
"learning_rate": 3.0500000000000004e-07,
|
| 85 |
+
"loss": 17.7823,
|
| 86 |
+
"step": 65
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.01,
|
| 90 |
+
"learning_rate": 3.3e-07,
|
| 91 |
+
"loss": 15.2998,
|
| 92 |
+
"step": 70
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.01,
|
| 96 |
+
"learning_rate": 3.55e-07,
|
| 97 |
+
"loss": 15.2838,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.01,
|
| 102 |
+
"learning_rate": 3.8e-07,
|
| 103 |
+
"loss": 14.8496,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.02,
|
| 108 |
+
"learning_rate": 4.0500000000000004e-07,
|
| 109 |
+
"loss": 14.3004,
|
| 110 |
+
"step": 85
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.02,
|
| 114 |
+
"learning_rate": 4.3e-07,
|
| 115 |
+
"loss": 14.2005,
|
| 116 |
+
"step": 90
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.02,
|
| 120 |
+
"learning_rate": 4.5500000000000004e-07,
|
| 121 |
+
"loss": 14.5054,
|
| 122 |
+
"step": 95
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.02,
|
| 126 |
+
"learning_rate": 4.800000000000001e-07,
|
| 127 |
+
"loss": 14.3888,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.02,
|
| 132 |
+
"learning_rate": 5.05e-07,
|
| 133 |
+
"loss": 17.3825,
|
| 134 |
+
"step": 105
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.02,
|
| 138 |
+
"learning_rate": 5.3e-07,
|
| 139 |
+
"loss": 17.0812,
|
| 140 |
+
"step": 110
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.02,
|
| 144 |
+
"learning_rate": 5.550000000000001e-07,
|
| 145 |
+
"loss": 17.0701,
|
| 146 |
+
"step": 115
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.02,
|
| 150 |
+
"learning_rate": 5.800000000000001e-07,
|
| 151 |
+
"loss": 14.9025,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.02,
|
| 156 |
+
"learning_rate": 6.05e-07,
|
| 157 |
+
"loss": 14.4225,
|
| 158 |
+
"step": 125
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.02,
|
| 162 |
+
"learning_rate": 6.3e-07,
|
| 163 |
+
"loss": 13.9425,
|
| 164 |
+
"step": 130
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.03,
|
| 168 |
+
"learning_rate": 6.550000000000001e-07,
|
| 169 |
+
"loss": 13.8037,
|
| 170 |
+
"step": 135
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.03,
|
| 174 |
+
"learning_rate": 6.800000000000001e-07,
|
| 175 |
+
"loss": 13.4372,
|
| 176 |
+
"step": 140
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.03,
|
| 180 |
+
"learning_rate": 7.05e-07,
|
| 181 |
+
"loss": 13.6807,
|
| 182 |
+
"step": 145
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.03,
|
| 186 |
+
"learning_rate": 7.3e-07,
|
| 187 |
+
"loss": 13.742,
|
| 188 |
+
"step": 150
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.03,
|
| 192 |
+
"learning_rate": 7.550000000000001e-07,
|
| 193 |
+
"loss": 15.1545,
|
| 194 |
+
"step": 155
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.03,
|
| 198 |
+
"learning_rate": 7.8e-07,
|
| 199 |
+
"loss": 16.1792,
|
| 200 |
+
"step": 160
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.03,
|
| 204 |
+
"learning_rate": 8.000000000000001e-07,
|
| 205 |
+
"loss": 14.6018,
|
| 206 |
+
"step": 165
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.03,
|
| 210 |
+
"learning_rate": 8.250000000000001e-07,
|
| 211 |
+
"loss": 12.6032,
|
| 212 |
+
"step": 170
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.03,
|
| 216 |
+
"learning_rate": 8.500000000000001e-07,
|
| 217 |
+
"loss": 13.0217,
|
| 218 |
+
"step": 175
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.03,
|
| 222 |
+
"learning_rate": 8.75e-07,
|
| 223 |
+
"loss": 12.2336,
|
| 224 |
+
"step": 180
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.03,
|
| 228 |
+
"learning_rate": 9.000000000000001e-07,
|
| 229 |
+
"loss": 11.8643,
|
| 230 |
+
"step": 185
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.04,
|
| 234 |
+
"learning_rate": 9.25e-07,
|
| 235 |
+
"loss": 11.8572,
|
| 236 |
+
"step": 190
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.04,
|
| 240 |
+
"learning_rate": 9.500000000000001e-07,
|
| 241 |
+
"loss": 10.8976,
|
| 242 |
+
"step": 195
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.04,
|
| 246 |
+
"learning_rate": 9.750000000000002e-07,
|
| 247 |
+
"loss": 12.2617,
|
| 248 |
+
"step": 200
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.04,
|
| 252 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 253 |
+
"loss": 13.7753,
|
| 254 |
+
"step": 205
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.04,
|
| 258 |
+
"learning_rate": 1.025e-06,
|
| 259 |
+
"loss": 11.6137,
|
| 260 |
+
"step": 210
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.04,
|
| 264 |
+
"learning_rate": 1.0500000000000001e-06,
|
| 265 |
+
"loss": 10.6402,
|
| 266 |
+
"step": 215
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.04,
|
| 270 |
+
"learning_rate": 1.075e-06,
|
| 271 |
+
"loss": 12.9591,
|
| 272 |
+
"step": 220
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.04,
|
| 276 |
+
"learning_rate": 1.1e-06,
|
| 277 |
+
"loss": 9.8613,
|
| 278 |
+
"step": 225
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.04,
|
| 282 |
+
"learning_rate": 1.125e-06,
|
| 283 |
+
"loss": 10.487,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.04,
|
| 288 |
+
"learning_rate": 1.1500000000000002e-06,
|
| 289 |
+
"loss": 9.3773,
|
| 290 |
+
"step": 235
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.04,
|
| 294 |
+
"learning_rate": 1.175e-06,
|
| 295 |
+
"loss": 9.5665,
|
| 296 |
+
"step": 240
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.05,
|
| 300 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 301 |
+
"loss": 8.2941,
|
| 302 |
+
"step": 245
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.05,
|
| 306 |
+
"learning_rate": 1.2250000000000001e-06,
|
| 307 |
+
"loss": 8.5563,
|
| 308 |
+
"step": 250
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.05,
|
| 312 |
+
"learning_rate": 1.25e-06,
|
| 313 |
+
"loss": 10.4941,
|
| 314 |
+
"step": 255
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.05,
|
| 318 |
+
"learning_rate": 1.275e-06,
|
| 319 |
+
"loss": 8.4986,
|
| 320 |
+
"step": 260
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.05,
|
| 324 |
+
"learning_rate": 1.3e-06,
|
| 325 |
+
"loss": 10.2295,
|
| 326 |
+
"step": 265
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.05,
|
| 330 |
+
"learning_rate": 1.3250000000000002e-06,
|
| 331 |
+
"loss": 8.7026,
|
| 332 |
+
"step": 270
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.05,
|
| 336 |
+
"learning_rate": 1.3500000000000002e-06,
|
| 337 |
+
"loss": 8.6009,
|
| 338 |
+
"step": 275
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.05,
|
| 342 |
+
"learning_rate": 1.3750000000000002e-06,
|
| 343 |
+
"loss": 7.7613,
|
| 344 |
+
"step": 280
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.05,
|
| 348 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 349 |
+
"loss": 7.8609,
|
| 350 |
+
"step": 285
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.05,
|
| 354 |
+
"learning_rate": 1.425e-06,
|
| 355 |
+
"loss": 7.0097,
|
| 356 |
+
"step": 290
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.05,
|
| 360 |
+
"learning_rate": 1.45e-06,
|
| 361 |
+
"loss": 5.5692,
|
| 362 |
+
"step": 295
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.06,
|
| 366 |
+
"learning_rate": 1.475e-06,
|
| 367 |
+
"loss": 5.6402,
|
| 368 |
+
"step": 300
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.06,
|
| 372 |
+
"learning_rate": 1.5e-06,
|
| 373 |
+
"loss": 9.0815,
|
| 374 |
+
"step": 305
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.06,
|
| 378 |
+
"learning_rate": 1.525e-06,
|
| 379 |
+
"loss": 8.0803,
|
| 380 |
+
"step": 310
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.06,
|
| 384 |
+
"learning_rate": 1.5500000000000002e-06,
|
| 385 |
+
"loss": 7.6229,
|
| 386 |
+
"step": 315
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.06,
|
| 390 |
+
"learning_rate": 1.5750000000000002e-06,
|
| 391 |
+
"loss": 5.5354,
|
| 392 |
+
"step": 320
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.06,
|
| 396 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 397 |
+
"loss": 8.1564,
|
| 398 |
+
"step": 325
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.06,
|
| 402 |
+
"learning_rate": 1.6250000000000001e-06,
|
| 403 |
+
"loss": 7.0378,
|
| 404 |
+
"step": 330
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.06,
|
| 408 |
+
"learning_rate": 1.6500000000000003e-06,
|
| 409 |
+
"loss": 5.3662,
|
| 410 |
+
"step": 335
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.06,
|
| 414 |
+
"learning_rate": 1.6750000000000003e-06,
|
| 415 |
+
"loss": 7.6436,
|
| 416 |
+
"step": 340
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.06,
|
| 420 |
+
"learning_rate": 1.7000000000000002e-06,
|
| 421 |
+
"loss": 5.3403,
|
| 422 |
+
"step": 345
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.07,
|
| 426 |
+
"learning_rate": 1.725e-06,
|
| 427 |
+
"loss": 8.1018,
|
| 428 |
+
"step": 350
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.07,
|
| 432 |
+
"learning_rate": 1.745e-06,
|
| 433 |
+
"loss": 7.5364,
|
| 434 |
+
"step": 355
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.07,
|
| 438 |
+
"learning_rate": 1.77e-06,
|
| 439 |
+
"loss": 5.2764,
|
| 440 |
+
"step": 360
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.07,
|
| 444 |
+
"learning_rate": 1.7950000000000002e-06,
|
| 445 |
+
"loss": 6.4889,
|
| 446 |
+
"step": 365
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.07,
|
| 450 |
+
"learning_rate": 1.8200000000000002e-06,
|
| 451 |
+
"loss": 5.1625,
|
| 452 |
+
"step": 370
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.07,
|
| 456 |
+
"learning_rate": 1.8450000000000001e-06,
|
| 457 |
+
"loss": 7.4417,
|
| 458 |
+
"step": 375
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.07,
|
| 462 |
+
"learning_rate": 1.87e-06,
|
| 463 |
+
"loss": 4.8924,
|
| 464 |
+
"step": 380
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.07,
|
| 468 |
+
"learning_rate": 1.895e-06,
|
| 469 |
+
"loss": 9.9698,
|
| 470 |
+
"step": 385
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.07,
|
| 474 |
+
"learning_rate": 1.9200000000000003e-06,
|
| 475 |
+
"loss": 7.4896,
|
| 476 |
+
"step": 390
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.07,
|
| 480 |
+
"learning_rate": 1.945e-06,
|
| 481 |
+
"loss": 6.7582,
|
| 482 |
+
"step": 395
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.07,
|
| 486 |
+
"learning_rate": 1.97e-06,
|
| 487 |
+
"loss": 6.0409,
|
| 488 |
+
"step": 400
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.08,
|
| 492 |
+
"learning_rate": 1.9950000000000004e-06,
|
| 493 |
+
"loss": 7.4142,
|
| 494 |
+
"step": 405
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.08,
|
| 498 |
+
"learning_rate": 2.02e-06,
|
| 499 |
+
"loss": 7.9007,
|
| 500 |
+
"step": 410
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.08,
|
| 504 |
+
"learning_rate": 2.045e-06,
|
| 505 |
+
"loss": 5.15,
|
| 506 |
+
"step": 415
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.08,
|
| 510 |
+
"learning_rate": 2.07e-06,
|
| 511 |
+
"loss": 5.6287,
|
| 512 |
+
"step": 420
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.08,
|
| 516 |
+
"learning_rate": 2.0950000000000003e-06,
|
| 517 |
+
"loss": 4.7277,
|
| 518 |
+
"step": 425
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.08,
|
| 522 |
+
"learning_rate": 2.12e-06,
|
| 523 |
+
"loss": 4.9955,
|
| 524 |
+
"step": 430
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.08,
|
| 528 |
+
"learning_rate": 2.1450000000000002e-06,
|
| 529 |
+
"loss": 6.4111,
|
| 530 |
+
"step": 435
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.08,
|
| 534 |
+
"learning_rate": 2.17e-06,
|
| 535 |
+
"loss": 5.7243,
|
| 536 |
+
"step": 440
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.08,
|
| 540 |
+
"learning_rate": 2.195e-06,
|
| 541 |
+
"loss": 5.3146,
|
| 542 |
+
"step": 445
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.08,
|
| 546 |
+
"learning_rate": 2.2200000000000003e-06,
|
| 547 |
+
"loss": 4.8272,
|
| 548 |
+
"step": 450
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.08,
|
| 552 |
+
"learning_rate": 2.245e-06,
|
| 553 |
+
"loss": 6.3851,
|
| 554 |
+
"step": 455
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.09,
|
| 558 |
+
"learning_rate": 2.2700000000000003e-06,
|
| 559 |
+
"loss": 5.0858,
|
| 560 |
+
"step": 460
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.09,
|
| 564 |
+
"learning_rate": 2.2950000000000005e-06,
|
| 565 |
+
"loss": 13.9773,
|
| 566 |
+
"step": 465
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.09,
|
| 570 |
+
"learning_rate": 2.3200000000000002e-06,
|
| 571 |
+
"loss": 5.1323,
|
| 572 |
+
"step": 470
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.09,
|
| 576 |
+
"learning_rate": 2.345e-06,
|
| 577 |
+
"loss": 4.5519,
|
| 578 |
+
"step": 475
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.09,
|
| 582 |
+
"learning_rate": 2.37e-06,
|
| 583 |
+
"loss": 4.9828,
|
| 584 |
+
"step": 480
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.09,
|
| 588 |
+
"learning_rate": 2.395e-06,
|
| 589 |
+
"loss": 10.9601,
|
| 590 |
+
"step": 485
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.09,
|
| 594 |
+
"learning_rate": 2.42e-06,
|
| 595 |
+
"loss": 4.715,
|
| 596 |
+
"step": 490
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.09,
|
| 600 |
+
"learning_rate": 2.4450000000000003e-06,
|
| 601 |
+
"loss": 6.1112,
|
| 602 |
+
"step": 495
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.09,
|
| 606 |
+
"learning_rate": 2.47e-06,
|
| 607 |
+
"loss": 4.8677,
|
| 608 |
+
"step": 500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.09,
|
| 612 |
+
"eval_loss": 6.384158611297607,
|
| 613 |
+
"eval_runtime": 97.6612,
|
| 614 |
+
"eval_samples_per_second": 12.226,
|
| 615 |
+
"eval_steps_per_second": 1.536,
|
| 616 |
+
"eval_wer": 1.0,
|
| 617 |
+
"step": 500
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.09,
|
| 621 |
+
"learning_rate": 2.4950000000000003e-06,
|
| 622 |
+
"loss": 5.6367,
|
| 623 |
+
"step": 505
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 0.09,
|
| 627 |
+
"learning_rate": 2.52e-06,
|
| 628 |
+
"loss": 4.798,
|
| 629 |
+
"step": 510
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.1,
|
| 633 |
+
"learning_rate": 2.545e-06,
|
| 634 |
+
"loss": 4.7321,
|
| 635 |
+
"step": 515
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.1,
|
| 639 |
+
"learning_rate": 2.5700000000000004e-06,
|
| 640 |
+
"loss": 9.0061,
|
| 641 |
+
"step": 520
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.1,
|
| 645 |
+
"learning_rate": 2.595e-06,
|
| 646 |
+
"loss": 4.5565,
|
| 647 |
+
"step": 525
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.1,
|
| 651 |
+
"learning_rate": 2.6200000000000003e-06,
|
| 652 |
+
"loss": 4.9119,
|
| 653 |
+
"step": 530
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.1,
|
| 657 |
+
"learning_rate": 2.6450000000000005e-06,
|
| 658 |
+
"loss": 4.6625,
|
| 659 |
+
"step": 535
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.1,
|
| 663 |
+
"learning_rate": 2.6700000000000003e-06,
|
| 664 |
+
"loss": 4.4122,
|
| 665 |
+
"step": 540
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.1,
|
| 669 |
+
"learning_rate": 2.6950000000000005e-06,
|
| 670 |
+
"loss": 6.515,
|
| 671 |
+
"step": 545
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.1,
|
| 675 |
+
"learning_rate": 2.7200000000000002e-06,
|
| 676 |
+
"loss": 6.2983,
|
| 677 |
+
"step": 550
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.1,
|
| 681 |
+
"learning_rate": 2.7450000000000004e-06,
|
| 682 |
+
"loss": 7.8164,
|
| 683 |
+
"step": 555
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.1,
|
| 687 |
+
"learning_rate": 2.7700000000000006e-06,
|
| 688 |
+
"loss": 5.5597,
|
| 689 |
+
"step": 560
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.11,
|
| 693 |
+
"learning_rate": 2.7950000000000003e-06,
|
| 694 |
+
"loss": 4.1535,
|
| 695 |
+
"step": 565
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.11,
|
| 699 |
+
"learning_rate": 2.82e-06,
|
| 700 |
+
"loss": 5.9433,
|
| 701 |
+
"step": 570
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 0.11,
|
| 705 |
+
"learning_rate": 2.845e-06,
|
| 706 |
+
"loss": 4.8775,
|
| 707 |
+
"step": 575
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 0.11,
|
| 711 |
+
"learning_rate": 2.87e-06,
|
| 712 |
+
"loss": 6.3474,
|
| 713 |
+
"step": 580
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.11,
|
| 717 |
+
"learning_rate": 2.8950000000000002e-06,
|
| 718 |
+
"loss": 8.6641,
|
| 719 |
+
"step": 585
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.11,
|
| 723 |
+
"learning_rate": 2.92e-06,
|
| 724 |
+
"loss": 6.0309,
|
| 725 |
+
"step": 590
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.11,
|
| 729 |
+
"learning_rate": 2.945e-06,
|
| 730 |
+
"loss": 4.1472,
|
| 731 |
+
"step": 595
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.11,
|
| 735 |
+
"learning_rate": 2.97e-06,
|
| 736 |
+
"loss": 4.5871,
|
| 737 |
+
"step": 600
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.11,
|
| 741 |
+
"learning_rate": 2.995e-06,
|
| 742 |
+
"loss": 5.0069,
|
| 743 |
+
"step": 605
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.11,
|
| 747 |
+
"learning_rate": 3.0200000000000003e-06,
|
| 748 |
+
"loss": 4.694,
|
| 749 |
+
"step": 610
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 0.11,
|
| 753 |
+
"learning_rate": 3.045e-06,
|
| 754 |
+
"loss": 4.502,
|
| 755 |
+
"step": 615
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 0.12,
|
| 759 |
+
"learning_rate": 3.0700000000000003e-06,
|
| 760 |
+
"loss": 7.2696,
|
| 761 |
+
"step": 620
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.12,
|
| 765 |
+
"learning_rate": 3.0950000000000004e-06,
|
| 766 |
+
"loss": 4.5793,
|
| 767 |
+
"step": 625
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.12,
|
| 771 |
+
"learning_rate": 3.12e-06,
|
| 772 |
+
"loss": 4.1937,
|
| 773 |
+
"step": 630
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.12,
|
| 777 |
+
"learning_rate": 3.1450000000000004e-06,
|
| 778 |
+
"loss": 6.3539,
|
| 779 |
+
"step": 635
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.12,
|
| 783 |
+
"learning_rate": 3.17e-06,
|
| 784 |
+
"loss": 4.1621,
|
| 785 |
+
"step": 640
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.12,
|
| 789 |
+
"learning_rate": 3.1950000000000003e-06,
|
| 790 |
+
"loss": 4.4824,
|
| 791 |
+
"step": 645
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 0.12,
|
| 795 |
+
"learning_rate": 3.2200000000000005e-06,
|
| 796 |
+
"loss": 4.4693,
|
| 797 |
+
"step": 650
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.12,
|
| 801 |
+
"learning_rate": 3.2450000000000003e-06,
|
| 802 |
+
"loss": 5.6584,
|
| 803 |
+
"step": 655
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.12,
|
| 807 |
+
"learning_rate": 3.2700000000000005e-06,
|
| 808 |
+
"loss": 4.813,
|
| 809 |
+
"step": 660
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.12,
|
| 813 |
+
"learning_rate": 3.2950000000000002e-06,
|
| 814 |
+
"loss": 6.1425,
|
| 815 |
+
"step": 665
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.12,
|
| 819 |
+
"learning_rate": 3.3200000000000004e-06,
|
| 820 |
+
"loss": 7.5331,
|
| 821 |
+
"step": 670
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.13,
|
| 825 |
+
"learning_rate": 3.3450000000000006e-06,
|
| 826 |
+
"loss": 9.772,
|
| 827 |
+
"step": 675
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.13,
|
| 831 |
+
"learning_rate": 3.3700000000000003e-06,
|
| 832 |
+
"loss": 4.1922,
|
| 833 |
+
"step": 680
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.13,
|
| 837 |
+
"learning_rate": 3.3950000000000005e-06,
|
| 838 |
+
"loss": 3.8468,
|
| 839 |
+
"step": 685
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 0.13,
|
| 843 |
+
"learning_rate": 3.4200000000000007e-06,
|
| 844 |
+
"loss": 3.5808,
|
| 845 |
+
"step": 690
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.13,
|
| 849 |
+
"learning_rate": 3.445e-06,
|
| 850 |
+
"loss": 5.5091,
|
| 851 |
+
"step": 695
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.13,
|
| 855 |
+
"learning_rate": 3.4700000000000002e-06,
|
| 856 |
+
"loss": 3.768,
|
| 857 |
+
"step": 700
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.13,
|
| 861 |
+
"learning_rate": 3.495e-06,
|
| 862 |
+
"loss": 4.9197,
|
| 863 |
+
"step": 705
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.13,
|
| 867 |
+
"learning_rate": 3.52e-06,
|
| 868 |
+
"loss": 4.1167,
|
| 869 |
+
"step": 710
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.13,
|
| 873 |
+
"learning_rate": 3.545e-06,
|
| 874 |
+
"loss": 7.1805,
|
| 875 |
+
"step": 715
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 0.13,
|
| 879 |
+
"learning_rate": 3.57e-06,
|
| 880 |
+
"loss": 7.7838,
|
| 881 |
+
"step": 720
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.14,
|
| 885 |
+
"learning_rate": 3.5950000000000003e-06,
|
| 886 |
+
"loss": 6.4851,
|
| 887 |
+
"step": 725
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.14,
|
| 891 |
+
"learning_rate": 3.62e-06,
|
| 892 |
+
"loss": 4.2357,
|
| 893 |
+
"step": 730
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 0.14,
|
| 897 |
+
"learning_rate": 3.6450000000000003e-06,
|
| 898 |
+
"loss": 5.7078,
|
| 899 |
+
"step": 735
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.14,
|
| 903 |
+
"learning_rate": 3.6700000000000004e-06,
|
| 904 |
+
"loss": 5.5984,
|
| 905 |
+
"step": 740
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.14,
|
| 909 |
+
"learning_rate": 3.695e-06,
|
| 910 |
+
"loss": 5.8425,
|
| 911 |
+
"step": 745
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.14,
|
| 915 |
+
"learning_rate": 3.7200000000000004e-06,
|
| 916 |
+
"loss": 3.9582,
|
| 917 |
+
"step": 750
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 0.14,
|
| 921 |
+
"learning_rate": 3.745e-06,
|
| 922 |
+
"loss": 6.1284,
|
| 923 |
+
"step": 755
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.14,
|
| 927 |
+
"learning_rate": 3.7700000000000003e-06,
|
| 928 |
+
"loss": 6.4174,
|
| 929 |
+
"step": 760
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.14,
|
| 933 |
+
"learning_rate": 3.7950000000000005e-06,
|
| 934 |
+
"loss": 4.0111,
|
| 935 |
+
"step": 765
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.14,
|
| 939 |
+
"learning_rate": 3.820000000000001e-06,
|
| 940 |
+
"loss": 6.0067,
|
| 941 |
+
"step": 770
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.14,
|
| 945 |
+
"learning_rate": 3.8450000000000005e-06,
|
| 946 |
+
"loss": 5.5284,
|
| 947 |
+
"step": 775
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.15,
|
| 951 |
+
"learning_rate": 3.87e-06,
|
| 952 |
+
"loss": 5.7256,
|
| 953 |
+
"step": 780
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.15,
|
| 957 |
+
"learning_rate": 3.895000000000001e-06,
|
| 958 |
+
"loss": 5.4687,
|
| 959 |
+
"step": 785
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.15,
|
| 963 |
+
"learning_rate": 3.920000000000001e-06,
|
| 964 |
+
"loss": 5.5711,
|
| 965 |
+
"step": 790
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 0.15,
|
| 969 |
+
"learning_rate": 3.945e-06,
|
| 970 |
+
"loss": 3.9495,
|
| 971 |
+
"step": 795
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 0.15,
|
| 975 |
+
"learning_rate": 3.97e-06,
|
| 976 |
+
"loss": 7.5433,
|
| 977 |
+
"step": 800
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.15,
|
| 981 |
+
"learning_rate": 3.995000000000001e-06,
|
| 982 |
+
"loss": 6.6596,
|
| 983 |
+
"step": 805
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.15,
|
| 987 |
+
"learning_rate": 4.0200000000000005e-06,
|
| 988 |
+
"loss": 3.8816,
|
| 989 |
+
"step": 810
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.15,
|
| 993 |
+
"learning_rate": 4.045e-06,
|
| 994 |
+
"loss": 6.0984,
|
| 995 |
+
"step": 815
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.15,
|
| 999 |
+
"learning_rate": 4.07e-06,
|
| 1000 |
+
"loss": 3.6398,
|
| 1001 |
+
"step": 820
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.15,
|
| 1005 |
+
"learning_rate": 4.095e-06,
|
| 1006 |
+
"loss": 6.2519,
|
| 1007 |
+
"step": 825
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 0.15,
|
| 1011 |
+
"learning_rate": 4.12e-06,
|
| 1012 |
+
"loss": 5.4348,
|
| 1013 |
+
"step": 830
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 0.16,
|
| 1017 |
+
"learning_rate": 4.145e-06,
|
| 1018 |
+
"loss": 5.7349,
|
| 1019 |
+
"step": 835
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 0.16,
|
| 1023 |
+
"learning_rate": 4.17e-06,
|
| 1024 |
+
"loss": 5.6213,
|
| 1025 |
+
"step": 840
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.16,
|
| 1029 |
+
"learning_rate": 4.1950000000000005e-06,
|
| 1030 |
+
"loss": 3.7084,
|
| 1031 |
+
"step": 845
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.16,
|
| 1035 |
+
"learning_rate": 4.22e-06,
|
| 1036 |
+
"loss": 7.2329,
|
| 1037 |
+
"step": 850
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.16,
|
| 1041 |
+
"learning_rate": 4.245e-06,
|
| 1042 |
+
"loss": 5.7584,
|
| 1043 |
+
"step": 855
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 0.16,
|
| 1047 |
+
"learning_rate": 4.270000000000001e-06,
|
| 1048 |
+
"loss": 6.9726,
|
| 1049 |
+
"step": 860
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.16,
|
| 1053 |
+
"learning_rate": 4.295e-06,
|
| 1054 |
+
"loss": 3.7044,
|
| 1055 |
+
"step": 865
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 0.16,
|
| 1059 |
+
"learning_rate": 4.32e-06,
|
| 1060 |
+
"loss": 4.0139,
|
| 1061 |
+
"step": 870
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.16,
|
| 1065 |
+
"learning_rate": 4.345000000000001e-06,
|
| 1066 |
+
"loss": 6.19,
|
| 1067 |
+
"step": 875
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.16,
|
| 1071 |
+
"learning_rate": 4.3700000000000005e-06,
|
| 1072 |
+
"loss": 6.5308,
|
| 1073 |
+
"step": 880
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.16,
|
| 1077 |
+
"learning_rate": 4.395e-06,
|
| 1078 |
+
"loss": 5.7261,
|
| 1079 |
+
"step": 885
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.17,
|
| 1083 |
+
"learning_rate": 4.42e-06,
|
| 1084 |
+
"loss": 3.8616,
|
| 1085 |
+
"step": 890
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 0.17,
|
| 1089 |
+
"learning_rate": 4.445000000000001e-06,
|
| 1090 |
+
"loss": 3.4939,
|
| 1091 |
+
"step": 895
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.17,
|
| 1095 |
+
"learning_rate": 4.47e-06,
|
| 1096 |
+
"loss": 3.5178,
|
| 1097 |
+
"step": 900
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 0.17,
|
| 1101 |
+
"learning_rate": 4.495e-06,
|
| 1102 |
+
"loss": 3.6419,
|
| 1103 |
+
"step": 905
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 0.17,
|
| 1107 |
+
"learning_rate": 4.520000000000001e-06,
|
| 1108 |
+
"loss": 8.8566,
|
| 1109 |
+
"step": 910
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.17,
|
| 1113 |
+
"learning_rate": 4.5450000000000005e-06,
|
| 1114 |
+
"loss": 3.7769,
|
| 1115 |
+
"step": 915
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.17,
|
| 1119 |
+
"learning_rate": 4.57e-06,
|
| 1120 |
+
"loss": 3.6874,
|
| 1121 |
+
"step": 920
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.17,
|
| 1125 |
+
"learning_rate": 4.595000000000001e-06,
|
| 1126 |
+
"loss": 5.7525,
|
| 1127 |
+
"step": 925
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 0.17,
|
| 1131 |
+
"learning_rate": 4.620000000000001e-06,
|
| 1132 |
+
"loss": 5.7784,
|
| 1133 |
+
"step": 930
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 0.17,
|
| 1137 |
+
"learning_rate": 4.645e-06,
|
| 1138 |
+
"loss": 5.0083,
|
| 1139 |
+
"step": 935
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 0.18,
|
| 1143 |
+
"learning_rate": 4.670000000000001e-06,
|
| 1144 |
+
"loss": 3.4095,
|
| 1145 |
+
"step": 940
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.18,
|
| 1149 |
+
"learning_rate": 4.695e-06,
|
| 1150 |
+
"loss": 3.212,
|
| 1151 |
+
"step": 945
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.18,
|
| 1155 |
+
"learning_rate": 4.7200000000000005e-06,
|
| 1156 |
+
"loss": 5.4597,
|
| 1157 |
+
"step": 950
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.18,
|
| 1161 |
+
"learning_rate": 4.745e-06,
|
| 1162 |
+
"loss": 3.7717,
|
| 1163 |
+
"step": 955
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 0.18,
|
| 1167 |
+
"learning_rate": 4.77e-06,
|
| 1168 |
+
"loss": 3.3286,
|
| 1169 |
+
"step": 960
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 0.18,
|
| 1173 |
+
"learning_rate": 4.795e-06,
|
| 1174 |
+
"loss": 3.3198,
|
| 1175 |
+
"step": 965
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 0.18,
|
| 1179 |
+
"learning_rate": 4.8200000000000004e-06,
|
| 1180 |
+
"loss": 3.6952,
|
| 1181 |
+
"step": 970
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 0.18,
|
| 1185 |
+
"learning_rate": 4.845e-06,
|
| 1186 |
+
"loss": 3.3483,
|
| 1187 |
+
"step": 975
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 0.18,
|
| 1191 |
+
"learning_rate": 4.87e-06,
|
| 1192 |
+
"loss": 4.8449,
|
| 1193 |
+
"step": 980
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.18,
|
| 1197 |
+
"learning_rate": 4.8950000000000006e-06,
|
| 1198 |
+
"loss": 3.5239,
|
| 1199 |
+
"step": 985
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.18,
|
| 1203 |
+
"learning_rate": 4.92e-06,
|
| 1204 |
+
"loss": 5.7172,
|
| 1205 |
+
"step": 990
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.19,
|
| 1209 |
+
"learning_rate": 4.945e-06,
|
| 1210 |
+
"loss": 3.9009,
|
| 1211 |
+
"step": 995
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 0.19,
|
| 1215 |
+
"learning_rate": 4.970000000000001e-06,
|
| 1216 |
+
"loss": 3.293,
|
| 1217 |
+
"step": 1000
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.19,
|
| 1221 |
+
"eval_loss": 5.126453876495361,
|
| 1222 |
+
"eval_runtime": 138.3538,
|
| 1223 |
+
"eval_samples_per_second": 8.63,
|
| 1224 |
+
"eval_steps_per_second": 1.084,
|
| 1225 |
+
"eval_wer": 1.0,
|
| 1226 |
+
"step": 1000
|
| 1227 |
+
}
|
| 1228 |
+
],
|
| 1229 |
+
"max_steps": 161070,
|
| 1230 |
+
"num_train_epochs": 30,
|
| 1231 |
+
"total_flos": 9.309229789594176e+16,
|
| 1232 |
+
"trial_name": null,
|
| 1233 |
+
"trial_params": null
|
| 1234 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:922e76dc3691c51d5f9343a92f5af52216f5d3cea75030305b0ccd89e28a94a7
|
| 3 |
+
size 2991
|
checkpoint-500/config.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-base-it-voxpopuli",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForCTC"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.1,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 256,
|
| 15 |
+
"contrastive_logits_temperature": 0.1,
|
| 16 |
+
"conv_bias": false,
|
| 17 |
+
"conv_dim": [
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512
|
| 25 |
+
],
|
| 26 |
+
"conv_kernel": [
|
| 27 |
+
10,
|
| 28 |
+
3,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
2,
|
| 33 |
+
2
|
| 34 |
+
],
|
| 35 |
+
"conv_stride": [
|
| 36 |
+
5,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"ctc_loss_reduction": "mean",
|
| 45 |
+
"ctc_zero_infinity": false,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": false,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "group",
|
| 52 |
+
"feat_proj_dropout": 0.1,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.1,
|
| 55 |
+
"gradient_checkpointing": false,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.1,
|
| 58 |
+
"hidden_dropout_prob": 0.1,
|
| 59 |
+
"hidden_size": 768,
|
| 60 |
+
"initializer_range": 0.02,
|
| 61 |
+
"intermediate_size": 3072,
|
| 62 |
+
"layer_norm_eps": 1e-05,
|
| 63 |
+
"layerdrop": 0.1,
|
| 64 |
+
"mask_feature_length": 10,
|
| 65 |
+
"mask_feature_min_masks": 0,
|
| 66 |
+
"mask_feature_prob": 0.0,
|
| 67 |
+
"mask_time_length": 10,
|
| 68 |
+
"mask_time_min_masks": 2,
|
| 69 |
+
"mask_time_prob": 0.05,
|
| 70 |
+
"model_type": "wav2vec2",
|
| 71 |
+
"num_adapter_layers": 3,
|
| 72 |
+
"num_attention_heads": 12,
|
| 73 |
+
"num_codevector_groups": 2,
|
| 74 |
+
"num_codevectors_per_group": 320,
|
| 75 |
+
"num_conv_pos_embedding_groups": 16,
|
| 76 |
+
"num_conv_pos_embeddings": 128,
|
| 77 |
+
"num_feat_extract_layers": 7,
|
| 78 |
+
"num_hidden_layers": 12,
|
| 79 |
+
"num_negatives": 100,
|
| 80 |
+
"output_hidden_size": 768,
|
| 81 |
+
"pad_token_id": 25,
|
| 82 |
+
"proj_codevector_dim": 256,
|
| 83 |
+
"tdnn_dilation": [
|
| 84 |
+
1,
|
| 85 |
+
2,
|
| 86 |
+
3,
|
| 87 |
+
1,
|
| 88 |
+
1
|
| 89 |
+
],
|
| 90 |
+
"tdnn_dim": [
|
| 91 |
+
512,
|
| 92 |
+
512,
|
| 93 |
+
512,
|
| 94 |
+
512,
|
| 95 |
+
1500
|
| 96 |
+
],
|
| 97 |
+
"tdnn_kernel": [
|
| 98 |
+
5,
|
| 99 |
+
3,
|
| 100 |
+
3,
|
| 101 |
+
1,
|
| 102 |
+
1
|
| 103 |
+
],
|
| 104 |
+
"torch_dtype": "float32",
|
| 105 |
+
"transformers_version": "4.16.1",
|
| 106 |
+
"use_weighted_layer_sum": false,
|
| 107 |
+
"vocab_size": 32,
|
| 108 |
+
"xvector_output_dim": 512
|
| 109 |
+
}
|
checkpoint-500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:805063f6f76295565aa2446ad59a4471544db1f20c8e3665c94833fd00aca33a
|
| 3 |
+
size 721685265
|
checkpoint-500/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": false,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
checkpoint-500/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:46317041606caab77c0ad60e704f2cc9a5a3d8fcd5be06e52d61b524fab2f16d
|
| 3 |
+
size 377670039
|
checkpoint-500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d175baed4be3fac4c9221252b16cc3e710a9904aa63cdbad7cdc197ffa7a3fcb
|
| 3 |
+
size 14567
|
checkpoint-500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e23af8c1bb1f8592677c1178bd4b2c7bc377906569378fcbd429440c1eb9f6f
|
| 3 |
+
size 559
|
checkpoint-500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ffd7f7416190754d4a242098b711847951ae6b96ee11f204d1d53a1c9d36e872
|
| 3 |
+
size 623
|
checkpoint-500/trainer_state.json
ADDED
|
@@ -0,0 +1,625 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.09312721177127957,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 1e-08,
|
| 13 |
+
"loss": 24.0683,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.0,
|
| 18 |
+
"learning_rate": 3.5e-08,
|
| 19 |
+
"loss": 17.9067,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.0,
|
| 24 |
+
"learning_rate": 6.000000000000001e-08,
|
| 25 |
+
"loss": 16.9004,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.0,
|
| 30 |
+
"learning_rate": 8.500000000000001e-08,
|
| 31 |
+
"loss": 16.8411,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.0,
|
| 36 |
+
"learning_rate": 1.1e-07,
|
| 37 |
+
"loss": 15.0697,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.01,
|
| 42 |
+
"learning_rate": 1.35e-07,
|
| 43 |
+
"loss": 15.2061,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01,
|
| 48 |
+
"learning_rate": 1.6e-07,
|
| 49 |
+
"loss": 14.8504,
|
| 50 |
+
"step": 35
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.01,
|
| 54 |
+
"learning_rate": 1.85e-07,
|
| 55 |
+
"loss": 14.6816,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.01,
|
| 60 |
+
"learning_rate": 2.1000000000000003e-07,
|
| 61 |
+
"loss": 14.9665,
|
| 62 |
+
"step": 45
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.01,
|
| 66 |
+
"learning_rate": 2.3500000000000003e-07,
|
| 67 |
+
"loss": 14.3427,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.01,
|
| 72 |
+
"learning_rate": 2.6e-07,
|
| 73 |
+
"loss": 18.4574,
|
| 74 |
+
"step": 55
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.01,
|
| 78 |
+
"learning_rate": 2.8e-07,
|
| 79 |
+
"loss": 18.379,
|
| 80 |
+
"step": 60
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.01,
|
| 84 |
+
"learning_rate": 3.0500000000000004e-07,
|
| 85 |
+
"loss": 17.7823,
|
| 86 |
+
"step": 65
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.01,
|
| 90 |
+
"learning_rate": 3.3e-07,
|
| 91 |
+
"loss": 15.2998,
|
| 92 |
+
"step": 70
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.01,
|
| 96 |
+
"learning_rate": 3.55e-07,
|
| 97 |
+
"loss": 15.2838,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.01,
|
| 102 |
+
"learning_rate": 3.8e-07,
|
| 103 |
+
"loss": 14.8496,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.02,
|
| 108 |
+
"learning_rate": 4.0500000000000004e-07,
|
| 109 |
+
"loss": 14.3004,
|
| 110 |
+
"step": 85
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.02,
|
| 114 |
+
"learning_rate": 4.3e-07,
|
| 115 |
+
"loss": 14.2005,
|
| 116 |
+
"step": 90
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.02,
|
| 120 |
+
"learning_rate": 4.5500000000000004e-07,
|
| 121 |
+
"loss": 14.5054,
|
| 122 |
+
"step": 95
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.02,
|
| 126 |
+
"learning_rate": 4.800000000000001e-07,
|
| 127 |
+
"loss": 14.3888,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.02,
|
| 132 |
+
"learning_rate": 5.05e-07,
|
| 133 |
+
"loss": 17.3825,
|
| 134 |
+
"step": 105
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.02,
|
| 138 |
+
"learning_rate": 5.3e-07,
|
| 139 |
+
"loss": 17.0812,
|
| 140 |
+
"step": 110
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.02,
|
| 144 |
+
"learning_rate": 5.550000000000001e-07,
|
| 145 |
+
"loss": 17.0701,
|
| 146 |
+
"step": 115
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.02,
|
| 150 |
+
"learning_rate": 5.800000000000001e-07,
|
| 151 |
+
"loss": 14.9025,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.02,
|
| 156 |
+
"learning_rate": 6.05e-07,
|
| 157 |
+
"loss": 14.4225,
|
| 158 |
+
"step": 125
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.02,
|
| 162 |
+
"learning_rate": 6.3e-07,
|
| 163 |
+
"loss": 13.9425,
|
| 164 |
+
"step": 130
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.03,
|
| 168 |
+
"learning_rate": 6.550000000000001e-07,
|
| 169 |
+
"loss": 13.8037,
|
| 170 |
+
"step": 135
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.03,
|
| 174 |
+
"learning_rate": 6.800000000000001e-07,
|
| 175 |
+
"loss": 13.4372,
|
| 176 |
+
"step": 140
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.03,
|
| 180 |
+
"learning_rate": 7.05e-07,
|
| 181 |
+
"loss": 13.6807,
|
| 182 |
+
"step": 145
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.03,
|
| 186 |
+
"learning_rate": 7.3e-07,
|
| 187 |
+
"loss": 13.742,
|
| 188 |
+
"step": 150
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.03,
|
| 192 |
+
"learning_rate": 7.550000000000001e-07,
|
| 193 |
+
"loss": 15.1545,
|
| 194 |
+
"step": 155
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.03,
|
| 198 |
+
"learning_rate": 7.8e-07,
|
| 199 |
+
"loss": 16.1792,
|
| 200 |
+
"step": 160
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.03,
|
| 204 |
+
"learning_rate": 8.000000000000001e-07,
|
| 205 |
+
"loss": 14.6018,
|
| 206 |
+
"step": 165
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.03,
|
| 210 |
+
"learning_rate": 8.250000000000001e-07,
|
| 211 |
+
"loss": 12.6032,
|
| 212 |
+
"step": 170
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.03,
|
| 216 |
+
"learning_rate": 8.500000000000001e-07,
|
| 217 |
+
"loss": 13.0217,
|
| 218 |
+
"step": 175
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.03,
|
| 222 |
+
"learning_rate": 8.75e-07,
|
| 223 |
+
"loss": 12.2336,
|
| 224 |
+
"step": 180
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.03,
|
| 228 |
+
"learning_rate": 9.000000000000001e-07,
|
| 229 |
+
"loss": 11.8643,
|
| 230 |
+
"step": 185
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.04,
|
| 234 |
+
"learning_rate": 9.25e-07,
|
| 235 |
+
"loss": 11.8572,
|
| 236 |
+
"step": 190
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.04,
|
| 240 |
+
"learning_rate": 9.500000000000001e-07,
|
| 241 |
+
"loss": 10.8976,
|
| 242 |
+
"step": 195
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.04,
|
| 246 |
+
"learning_rate": 9.750000000000002e-07,
|
| 247 |
+
"loss": 12.2617,
|
| 248 |
+
"step": 200
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.04,
|
| 252 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 253 |
+
"loss": 13.7753,
|
| 254 |
+
"step": 205
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.04,
|
| 258 |
+
"learning_rate": 1.025e-06,
|
| 259 |
+
"loss": 11.6137,
|
| 260 |
+
"step": 210
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.04,
|
| 264 |
+
"learning_rate": 1.0500000000000001e-06,
|
| 265 |
+
"loss": 10.6402,
|
| 266 |
+
"step": 215
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.04,
|
| 270 |
+
"learning_rate": 1.075e-06,
|
| 271 |
+
"loss": 12.9591,
|
| 272 |
+
"step": 220
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.04,
|
| 276 |
+
"learning_rate": 1.1e-06,
|
| 277 |
+
"loss": 9.8613,
|
| 278 |
+
"step": 225
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.04,
|
| 282 |
+
"learning_rate": 1.125e-06,
|
| 283 |
+
"loss": 10.487,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.04,
|
| 288 |
+
"learning_rate": 1.1500000000000002e-06,
|
| 289 |
+
"loss": 9.3773,
|
| 290 |
+
"step": 235
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.04,
|
| 294 |
+
"learning_rate": 1.175e-06,
|
| 295 |
+
"loss": 9.5665,
|
| 296 |
+
"step": 240
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.05,
|
| 300 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 301 |
+
"loss": 8.2941,
|
| 302 |
+
"step": 245
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.05,
|
| 306 |
+
"learning_rate": 1.2250000000000001e-06,
|
| 307 |
+
"loss": 8.5563,
|
| 308 |
+
"step": 250
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.05,
|
| 312 |
+
"learning_rate": 1.25e-06,
|
| 313 |
+
"loss": 10.4941,
|
| 314 |
+
"step": 255
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.05,
|
| 318 |
+
"learning_rate": 1.275e-06,
|
| 319 |
+
"loss": 8.4986,
|
| 320 |
+
"step": 260
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.05,
|
| 324 |
+
"learning_rate": 1.3e-06,
|
| 325 |
+
"loss": 10.2295,
|
| 326 |
+
"step": 265
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.05,
|
| 330 |
+
"learning_rate": 1.3250000000000002e-06,
|
| 331 |
+
"loss": 8.7026,
|
| 332 |
+
"step": 270
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.05,
|
| 336 |
+
"learning_rate": 1.3500000000000002e-06,
|
| 337 |
+
"loss": 8.6009,
|
| 338 |
+
"step": 275
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.05,
|
| 342 |
+
"learning_rate": 1.3750000000000002e-06,
|
| 343 |
+
"loss": 7.7613,
|
| 344 |
+
"step": 280
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.05,
|
| 348 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 349 |
+
"loss": 7.8609,
|
| 350 |
+
"step": 285
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.05,
|
| 354 |
+
"learning_rate": 1.425e-06,
|
| 355 |
+
"loss": 7.0097,
|
| 356 |
+
"step": 290
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.05,
|
| 360 |
+
"learning_rate": 1.45e-06,
|
| 361 |
+
"loss": 5.5692,
|
| 362 |
+
"step": 295
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.06,
|
| 366 |
+
"learning_rate": 1.475e-06,
|
| 367 |
+
"loss": 5.6402,
|
| 368 |
+
"step": 300
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.06,
|
| 372 |
+
"learning_rate": 1.5e-06,
|
| 373 |
+
"loss": 9.0815,
|
| 374 |
+
"step": 305
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.06,
|
| 378 |
+
"learning_rate": 1.525e-06,
|
| 379 |
+
"loss": 8.0803,
|
| 380 |
+
"step": 310
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.06,
|
| 384 |
+
"learning_rate": 1.5500000000000002e-06,
|
| 385 |
+
"loss": 7.6229,
|
| 386 |
+
"step": 315
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.06,
|
| 390 |
+
"learning_rate": 1.5750000000000002e-06,
|
| 391 |
+
"loss": 5.5354,
|
| 392 |
+
"step": 320
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.06,
|
| 396 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 397 |
+
"loss": 8.1564,
|
| 398 |
+
"step": 325
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.06,
|
| 402 |
+
"learning_rate": 1.6250000000000001e-06,
|
| 403 |
+
"loss": 7.0378,
|
| 404 |
+
"step": 330
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.06,
|
| 408 |
+
"learning_rate": 1.6500000000000003e-06,
|
| 409 |
+
"loss": 5.3662,
|
| 410 |
+
"step": 335
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.06,
|
| 414 |
+
"learning_rate": 1.6750000000000003e-06,
|
| 415 |
+
"loss": 7.6436,
|
| 416 |
+
"step": 340
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.06,
|
| 420 |
+
"learning_rate": 1.7000000000000002e-06,
|
| 421 |
+
"loss": 5.3403,
|
| 422 |
+
"step": 345
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.07,
|
| 426 |
+
"learning_rate": 1.725e-06,
|
| 427 |
+
"loss": 8.1018,
|
| 428 |
+
"step": 350
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.07,
|
| 432 |
+
"learning_rate": 1.745e-06,
|
| 433 |
+
"loss": 7.5364,
|
| 434 |
+
"step": 355
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.07,
|
| 438 |
+
"learning_rate": 1.77e-06,
|
| 439 |
+
"loss": 5.2764,
|
| 440 |
+
"step": 360
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.07,
|
| 444 |
+
"learning_rate": 1.7950000000000002e-06,
|
| 445 |
+
"loss": 6.4889,
|
| 446 |
+
"step": 365
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.07,
|
| 450 |
+
"learning_rate": 1.8200000000000002e-06,
|
| 451 |
+
"loss": 5.1625,
|
| 452 |
+
"step": 370
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.07,
|
| 456 |
+
"learning_rate": 1.8450000000000001e-06,
|
| 457 |
+
"loss": 7.4417,
|
| 458 |
+
"step": 375
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.07,
|
| 462 |
+
"learning_rate": 1.87e-06,
|
| 463 |
+
"loss": 4.8924,
|
| 464 |
+
"step": 380
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.07,
|
| 468 |
+
"learning_rate": 1.895e-06,
|
| 469 |
+
"loss": 9.9698,
|
| 470 |
+
"step": 385
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.07,
|
| 474 |
+
"learning_rate": 1.9200000000000003e-06,
|
| 475 |
+
"loss": 7.4896,
|
| 476 |
+
"step": 390
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.07,
|
| 480 |
+
"learning_rate": 1.945e-06,
|
| 481 |
+
"loss": 6.7582,
|
| 482 |
+
"step": 395
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.07,
|
| 486 |
+
"learning_rate": 1.97e-06,
|
| 487 |
+
"loss": 6.0409,
|
| 488 |
+
"step": 400
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.08,
|
| 492 |
+
"learning_rate": 1.9950000000000004e-06,
|
| 493 |
+
"loss": 7.4142,
|
| 494 |
+
"step": 405
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.08,
|
| 498 |
+
"learning_rate": 2.02e-06,
|
| 499 |
+
"loss": 7.9007,
|
| 500 |
+
"step": 410
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.08,
|
| 504 |
+
"learning_rate": 2.045e-06,
|
| 505 |
+
"loss": 5.15,
|
| 506 |
+
"step": 415
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.08,
|
| 510 |
+
"learning_rate": 2.07e-06,
|
| 511 |
+
"loss": 5.6287,
|
| 512 |
+
"step": 420
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.08,
|
| 516 |
+
"learning_rate": 2.0950000000000003e-06,
|
| 517 |
+
"loss": 4.7277,
|
| 518 |
+
"step": 425
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.08,
|
| 522 |
+
"learning_rate": 2.12e-06,
|
| 523 |
+
"loss": 4.9955,
|
| 524 |
+
"step": 430
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.08,
|
| 528 |
+
"learning_rate": 2.1450000000000002e-06,
|
| 529 |
+
"loss": 6.4111,
|
| 530 |
+
"step": 435
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.08,
|
| 534 |
+
"learning_rate": 2.17e-06,
|
| 535 |
+
"loss": 5.7243,
|
| 536 |
+
"step": 440
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.08,
|
| 540 |
+
"learning_rate": 2.195e-06,
|
| 541 |
+
"loss": 5.3146,
|
| 542 |
+
"step": 445
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.08,
|
| 546 |
+
"learning_rate": 2.2200000000000003e-06,
|
| 547 |
+
"loss": 4.8272,
|
| 548 |
+
"step": 450
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.08,
|
| 552 |
+
"learning_rate": 2.245e-06,
|
| 553 |
+
"loss": 6.3851,
|
| 554 |
+
"step": 455
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.09,
|
| 558 |
+
"learning_rate": 2.2700000000000003e-06,
|
| 559 |
+
"loss": 5.0858,
|
| 560 |
+
"step": 460
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.09,
|
| 564 |
+
"learning_rate": 2.2950000000000005e-06,
|
| 565 |
+
"loss": 13.9773,
|
| 566 |
+
"step": 465
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.09,
|
| 570 |
+
"learning_rate": 2.3200000000000002e-06,
|
| 571 |
+
"loss": 5.1323,
|
| 572 |
+
"step": 470
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.09,
|
| 576 |
+
"learning_rate": 2.345e-06,
|
| 577 |
+
"loss": 4.5519,
|
| 578 |
+
"step": 475
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.09,
|
| 582 |
+
"learning_rate": 2.37e-06,
|
| 583 |
+
"loss": 4.9828,
|
| 584 |
+
"step": 480
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.09,
|
| 588 |
+
"learning_rate": 2.395e-06,
|
| 589 |
+
"loss": 10.9601,
|
| 590 |
+
"step": 485
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.09,
|
| 594 |
+
"learning_rate": 2.42e-06,
|
| 595 |
+
"loss": 4.715,
|
| 596 |
+
"step": 490
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.09,
|
| 600 |
+
"learning_rate": 2.4450000000000003e-06,
|
| 601 |
+
"loss": 6.1112,
|
| 602 |
+
"step": 495
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.09,
|
| 606 |
+
"learning_rate": 2.47e-06,
|
| 607 |
+
"loss": 4.8677,
|
| 608 |
+
"step": 500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.09,
|
| 612 |
+
"eval_loss": 6.384158611297607,
|
| 613 |
+
"eval_runtime": 97.6612,
|
| 614 |
+
"eval_samples_per_second": 12.226,
|
| 615 |
+
"eval_steps_per_second": 1.536,
|
| 616 |
+
"eval_wer": 1.0,
|
| 617 |
+
"step": 500
|
| 618 |
+
}
|
| 619 |
+
],
|
| 620 |
+
"max_steps": 161070,
|
| 621 |
+
"num_train_epochs": 30,
|
| 622 |
+
"total_flos": 4.640032532404224e+16,
|
| 623 |
+
"trial_name": null,
|
| 624 |
+
"trial_params": null
|
| 625 |
+
}
|
checkpoint-500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:922e76dc3691c51d5f9343a92f5af52216f5d3cea75030305b0ccd89e28a94a7
|
| 3 |
+
size 2991
|