Training in progress, step 1170, checkpoint
Browse files- last-checkpoint/config.json +25 -0
- last-checkpoint/generation_config.json +6 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state_0.pth +3 -0
- last-checkpoint/rng_state_1.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/special_tokens_map.json +23 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer.model +3 -0
- last-checkpoint/tokenizer_config.json +42 -0
- last-checkpoint/trainer_state.json +2065 -0
- last-checkpoint/training_args.bin +3 -0
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MistralForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"bos_token_id": 1,
|
| 7 |
+
"eos_token_id": 2,
|
| 8 |
+
"hidden_act": "silu",
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 3072,
|
| 12 |
+
"max_position_embeddings": 32768,
|
| 13 |
+
"model_type": "mistral",
|
| 14 |
+
"num_attention_heads": 12,
|
| 15 |
+
"num_hidden_layers": 12,
|
| 16 |
+
"num_key_value_heads": 12,
|
| 17 |
+
"rms_norm_eps": 1e-06,
|
| 18 |
+
"rope_theta": 10000.0,
|
| 19 |
+
"sliding_window": 4096,
|
| 20 |
+
"tie_word_embeddings": false,
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.39.0.dev0",
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 32000
|
| 25 |
+
}
|
last-checkpoint/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.39.0.dev0"
|
| 6 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94d09c088c0ee76dfd5bc1256103385e076dc588cac73e73ac4b49395a1b092b
|
| 3 |
+
size 649681952
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fd891cdd54fea99b620e990f75d2715f4767b5cd0169f08d089aab20ee8987a
|
| 3 |
+
size 1299433082
|
last-checkpoint/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4f55eccb7975b84e9a3ab402dd2d6ad00e5aec7426982ecbe92ed0958c0926e
|
| 3 |
+
size 14512
|
last-checkpoint/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e5edeef57189b387f7bb9e6a2f50b9af301a6c5d8472047a0a665624715b5f4
|
| 3 |
+
size 14512
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a793dd3d5466229a639af9c1a0cc36cccb60dcd8adaa75d9cb8accecdabfa1f
|
| 3 |
+
size 1064
|
last-checkpoint/special_tokens_map.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"unk_token": {
|
| 17 |
+
"content": "<unk>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
}
|
| 23 |
+
}
|
last-checkpoint/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
last-checkpoint/tokenizer_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<unk>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"additional_special_tokens": [],
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "</s>",
|
| 34 |
+
"legacy": true,
|
| 35 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 36 |
+
"pad_token": null,
|
| 37 |
+
"sp_model_kwargs": {},
|
| 38 |
+
"spaces_between_special_tokens": false,
|
| 39 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 40 |
+
"unk_token": "<unk>",
|
| 41 |
+
"use_default_system_prompt": false
|
| 42 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,2065 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.10003419972640219,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1170,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"grad_norm": 4.3728413581848145,
|
| 14 |
+
"learning_rate": 4.998290013679891e-05,
|
| 15 |
+
"loss": 10.0117,
|
| 16 |
+
"step": 4
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0,
|
| 20 |
+
"grad_norm": 2.599367380142212,
|
| 21 |
+
"learning_rate": 4.996580027359781e-05,
|
| 22 |
+
"loss": 9.1745,
|
| 23 |
+
"step": 8
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0,
|
| 27 |
+
"grad_norm": 2.4951720237731934,
|
| 28 |
+
"learning_rate": 4.994870041039672e-05,
|
| 29 |
+
"loss": 8.7624,
|
| 30 |
+
"step": 12
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0,
|
| 34 |
+
"grad_norm": 2.4932913780212402,
|
| 35 |
+
"learning_rate": 4.9931600547195625e-05,
|
| 36 |
+
"loss": 8.5709,
|
| 37 |
+
"step": 16
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0,
|
| 41 |
+
"grad_norm": 2.3983781337738037,
|
| 42 |
+
"learning_rate": 4.991450068399453e-05,
|
| 43 |
+
"loss": 8.4411,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0,
|
| 48 |
+
"grad_norm": 2.2340972423553467,
|
| 49 |
+
"learning_rate": 4.989740082079344e-05,
|
| 50 |
+
"loss": 8.1957,
|
| 51 |
+
"step": 24
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0,
|
| 55 |
+
"grad_norm": 2.1964590549468994,
|
| 56 |
+
"learning_rate": 4.988030095759234e-05,
|
| 57 |
+
"loss": 8.0598,
|
| 58 |
+
"step": 28
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0,
|
| 62 |
+
"grad_norm": 1.9686617851257324,
|
| 63 |
+
"learning_rate": 4.986320109439125e-05,
|
| 64 |
+
"loss": 7.8679,
|
| 65 |
+
"step": 32
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0,
|
| 69 |
+
"grad_norm": 1.8235087394714355,
|
| 70 |
+
"learning_rate": 4.984610123119015e-05,
|
| 71 |
+
"loss": 7.7621,
|
| 72 |
+
"step": 36
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0,
|
| 76 |
+
"grad_norm": 1.7598885297775269,
|
| 77 |
+
"learning_rate": 4.9829001367989056e-05,
|
| 78 |
+
"loss": 7.5417,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0,
|
| 83 |
+
"grad_norm": 1.94789719581604,
|
| 84 |
+
"learning_rate": 4.981190150478797e-05,
|
| 85 |
+
"loss": 7.3331,
|
| 86 |
+
"step": 44
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0,
|
| 90 |
+
"grad_norm": 1.9366331100463867,
|
| 91 |
+
"learning_rate": 4.979480164158687e-05,
|
| 92 |
+
"loss": 7.1039,
|
| 93 |
+
"step": 48
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0,
|
| 97 |
+
"grad_norm": 2.1228675842285156,
|
| 98 |
+
"learning_rate": 4.977770177838578e-05,
|
| 99 |
+
"loss": 6.9965,
|
| 100 |
+
"step": 52
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0,
|
| 104 |
+
"grad_norm": 1.9855167865753174,
|
| 105 |
+
"learning_rate": 4.976060191518468e-05,
|
| 106 |
+
"loss": 6.8865,
|
| 107 |
+
"step": 56
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.01,
|
| 111 |
+
"grad_norm": 1.8035985231399536,
|
| 112 |
+
"learning_rate": 4.9743502051983585e-05,
|
| 113 |
+
"loss": 6.6322,
|
| 114 |
+
"step": 60
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.01,
|
| 118 |
+
"grad_norm": 1.8589977025985718,
|
| 119 |
+
"learning_rate": 4.9726402188782486e-05,
|
| 120 |
+
"loss": 6.5616,
|
| 121 |
+
"step": 64
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.01,
|
| 125 |
+
"grad_norm": 1.745139241218567,
|
| 126 |
+
"learning_rate": 4.97093023255814e-05,
|
| 127 |
+
"loss": 6.3479,
|
| 128 |
+
"step": 68
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.01,
|
| 132 |
+
"grad_norm": 1.266196608543396,
|
| 133 |
+
"learning_rate": 4.969220246238031e-05,
|
| 134 |
+
"loss": 6.2343,
|
| 135 |
+
"step": 72
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.01,
|
| 139 |
+
"grad_norm": 2.005223035812378,
|
| 140 |
+
"learning_rate": 4.967510259917921e-05,
|
| 141 |
+
"loss": 6.1866,
|
| 142 |
+
"step": 76
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.01,
|
| 146 |
+
"grad_norm": 1.5017377138137817,
|
| 147 |
+
"learning_rate": 4.9658002735978115e-05,
|
| 148 |
+
"loss": 5.972,
|
| 149 |
+
"step": 80
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.01,
|
| 153 |
+
"grad_norm": 1.6136974096298218,
|
| 154 |
+
"learning_rate": 4.9640902872777016e-05,
|
| 155 |
+
"loss": 6.1335,
|
| 156 |
+
"step": 84
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.01,
|
| 160 |
+
"grad_norm": 1.3865970373153687,
|
| 161 |
+
"learning_rate": 4.962380300957592e-05,
|
| 162 |
+
"loss": 5.944,
|
| 163 |
+
"step": 88
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.01,
|
| 167 |
+
"grad_norm": 1.283933162689209,
|
| 168 |
+
"learning_rate": 4.960670314637483e-05,
|
| 169 |
+
"loss": 5.8713,
|
| 170 |
+
"step": 92
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.01,
|
| 174 |
+
"grad_norm": 1.1588549613952637,
|
| 175 |
+
"learning_rate": 4.958960328317374e-05,
|
| 176 |
+
"loss": 5.7731,
|
| 177 |
+
"step": 96
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.01,
|
| 181 |
+
"grad_norm": 1.3081687688827515,
|
| 182 |
+
"learning_rate": 4.9572503419972645e-05,
|
| 183 |
+
"loss": 5.7429,
|
| 184 |
+
"step": 100
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.01,
|
| 188 |
+
"grad_norm": 1.6212745904922485,
|
| 189 |
+
"learning_rate": 4.9555403556771546e-05,
|
| 190 |
+
"loss": 5.481,
|
| 191 |
+
"step": 104
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.01,
|
| 195 |
+
"grad_norm": 1.709843397140503,
|
| 196 |
+
"learning_rate": 4.953830369357045e-05,
|
| 197 |
+
"loss": 5.4855,
|
| 198 |
+
"step": 108
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.01,
|
| 202 |
+
"grad_norm": 1.2814812660217285,
|
| 203 |
+
"learning_rate": 4.952120383036936e-05,
|
| 204 |
+
"loss": 5.4367,
|
| 205 |
+
"step": 112
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.01,
|
| 209 |
+
"grad_norm": 1.5539968013763428,
|
| 210 |
+
"learning_rate": 4.950410396716826e-05,
|
| 211 |
+
"loss": 5.2652,
|
| 212 |
+
"step": 116
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.01,
|
| 216 |
+
"grad_norm": 1.500375509262085,
|
| 217 |
+
"learning_rate": 4.948700410396717e-05,
|
| 218 |
+
"loss": 5.5137,
|
| 219 |
+
"step": 120
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.01,
|
| 223 |
+
"grad_norm": 1.7938899993896484,
|
| 224 |
+
"learning_rate": 4.9469904240766076e-05,
|
| 225 |
+
"loss": 5.4663,
|
| 226 |
+
"step": 124
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.01,
|
| 230 |
+
"grad_norm": 1.6169573068618774,
|
| 231 |
+
"learning_rate": 4.945280437756498e-05,
|
| 232 |
+
"loss": 5.3137,
|
| 233 |
+
"step": 128
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.01,
|
| 237 |
+
"grad_norm": 1.1904114484786987,
|
| 238 |
+
"learning_rate": 4.943570451436389e-05,
|
| 239 |
+
"loss": 5.1871,
|
| 240 |
+
"step": 132
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.01,
|
| 244 |
+
"grad_norm": 2.5564723014831543,
|
| 245 |
+
"learning_rate": 4.941860465116279e-05,
|
| 246 |
+
"loss": 5.4058,
|
| 247 |
+
"step": 136
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.01,
|
| 251 |
+
"grad_norm": 1.6187268495559692,
|
| 252 |
+
"learning_rate": 4.94015047879617e-05,
|
| 253 |
+
"loss": 5.1919,
|
| 254 |
+
"step": 140
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.01,
|
| 258 |
+
"grad_norm": 1.2222367525100708,
|
| 259 |
+
"learning_rate": 4.93844049247606e-05,
|
| 260 |
+
"loss": 5.1876,
|
| 261 |
+
"step": 144
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.01,
|
| 265 |
+
"grad_norm": 1.5898586511611938,
|
| 266 |
+
"learning_rate": 4.936730506155951e-05,
|
| 267 |
+
"loss": 5.1217,
|
| 268 |
+
"step": 148
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.01,
|
| 272 |
+
"grad_norm": 1.446902871131897,
|
| 273 |
+
"learning_rate": 4.935020519835842e-05,
|
| 274 |
+
"loss": 5.0582,
|
| 275 |
+
"step": 152
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.01,
|
| 279 |
+
"grad_norm": 1.862309217453003,
|
| 280 |
+
"learning_rate": 4.933310533515732e-05,
|
| 281 |
+
"loss": 4.9028,
|
| 282 |
+
"step": 156
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.01,
|
| 286 |
+
"grad_norm": 1.5455704927444458,
|
| 287 |
+
"learning_rate": 4.931600547195623e-05,
|
| 288 |
+
"loss": 5.0352,
|
| 289 |
+
"step": 160
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.01,
|
| 293 |
+
"grad_norm": 1.772558569908142,
|
| 294 |
+
"learning_rate": 4.929890560875513e-05,
|
| 295 |
+
"loss": 4.9707,
|
| 296 |
+
"step": 164
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.01,
|
| 300 |
+
"grad_norm": 1.8154480457305908,
|
| 301 |
+
"learning_rate": 4.9281805745554036e-05,
|
| 302 |
+
"loss": 5.0588,
|
| 303 |
+
"step": 168
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.01,
|
| 307 |
+
"grad_norm": 1.4536504745483398,
|
| 308 |
+
"learning_rate": 4.9264705882352944e-05,
|
| 309 |
+
"loss": 4.9997,
|
| 310 |
+
"step": 172
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.02,
|
| 314 |
+
"grad_norm": 1.6166527271270752,
|
| 315 |
+
"learning_rate": 4.924760601915185e-05,
|
| 316 |
+
"loss": 4.9472,
|
| 317 |
+
"step": 176
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.02,
|
| 321 |
+
"grad_norm": 2.1260979175567627,
|
| 322 |
+
"learning_rate": 4.923050615595076e-05,
|
| 323 |
+
"loss": 4.944,
|
| 324 |
+
"step": 180
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.02,
|
| 328 |
+
"grad_norm": 1.4778213500976562,
|
| 329 |
+
"learning_rate": 4.921340629274966e-05,
|
| 330 |
+
"loss": 4.8657,
|
| 331 |
+
"step": 184
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.02,
|
| 335 |
+
"grad_norm": 1.255488395690918,
|
| 336 |
+
"learning_rate": 4.9196306429548566e-05,
|
| 337 |
+
"loss": 4.7728,
|
| 338 |
+
"step": 188
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.02,
|
| 342 |
+
"grad_norm": 1.5698314905166626,
|
| 343 |
+
"learning_rate": 4.917920656634747e-05,
|
| 344 |
+
"loss": 4.9433,
|
| 345 |
+
"step": 192
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.02,
|
| 349 |
+
"grad_norm": 1.342402696609497,
|
| 350 |
+
"learning_rate": 4.9162106703146374e-05,
|
| 351 |
+
"loss": 4.6336,
|
| 352 |
+
"step": 196
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.02,
|
| 356 |
+
"grad_norm": 1.4427978992462158,
|
| 357 |
+
"learning_rate": 4.914500683994528e-05,
|
| 358 |
+
"loss": 4.7112,
|
| 359 |
+
"step": 200
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.02,
|
| 363 |
+
"grad_norm": 1.6528425216674805,
|
| 364 |
+
"learning_rate": 4.912790697674419e-05,
|
| 365 |
+
"loss": 4.876,
|
| 366 |
+
"step": 204
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.02,
|
| 370 |
+
"grad_norm": 1.706386923789978,
|
| 371 |
+
"learning_rate": 4.9110807113543096e-05,
|
| 372 |
+
"loss": 4.5971,
|
| 373 |
+
"step": 208
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.02,
|
| 377 |
+
"grad_norm": 1.7773737907409668,
|
| 378 |
+
"learning_rate": 4.9093707250342e-05,
|
| 379 |
+
"loss": 4.7944,
|
| 380 |
+
"step": 212
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.02,
|
| 384 |
+
"grad_norm": 1.7884222269058228,
|
| 385 |
+
"learning_rate": 4.9076607387140904e-05,
|
| 386 |
+
"loss": 4.5912,
|
| 387 |
+
"step": 216
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.02,
|
| 391 |
+
"grad_norm": 1.5389587879180908,
|
| 392 |
+
"learning_rate": 4.905950752393981e-05,
|
| 393 |
+
"loss": 4.6656,
|
| 394 |
+
"step": 220
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.02,
|
| 398 |
+
"grad_norm": 1.7006316184997559,
|
| 399 |
+
"learning_rate": 4.904240766073871e-05,
|
| 400 |
+
"loss": 4.6589,
|
| 401 |
+
"step": 224
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.02,
|
| 405 |
+
"grad_norm": 1.6314700841903687,
|
| 406 |
+
"learning_rate": 4.9025307797537626e-05,
|
| 407 |
+
"loss": 4.5036,
|
| 408 |
+
"step": 228
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.02,
|
| 412 |
+
"grad_norm": 1.7512930631637573,
|
| 413 |
+
"learning_rate": 4.900820793433653e-05,
|
| 414 |
+
"loss": 4.5105,
|
| 415 |
+
"step": 232
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.02,
|
| 419 |
+
"grad_norm": 1.9109033346176147,
|
| 420 |
+
"learning_rate": 4.8991108071135434e-05,
|
| 421 |
+
"loss": 4.4449,
|
| 422 |
+
"step": 236
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.02,
|
| 426 |
+
"grad_norm": 1.547317385673523,
|
| 427 |
+
"learning_rate": 4.897400820793434e-05,
|
| 428 |
+
"loss": 4.4906,
|
| 429 |
+
"step": 240
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.02,
|
| 433 |
+
"grad_norm": 1.5143060684204102,
|
| 434 |
+
"learning_rate": 4.895690834473324e-05,
|
| 435 |
+
"loss": 4.3545,
|
| 436 |
+
"step": 244
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.02,
|
| 440 |
+
"grad_norm": 1.6648136377334595,
|
| 441 |
+
"learning_rate": 4.893980848153215e-05,
|
| 442 |
+
"loss": 4.4093,
|
| 443 |
+
"step": 248
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.02,
|
| 447 |
+
"grad_norm": 1.5027951002120972,
|
| 448 |
+
"learning_rate": 4.892270861833106e-05,
|
| 449 |
+
"loss": 4.3444,
|
| 450 |
+
"step": 252
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.02,
|
| 454 |
+
"grad_norm": 1.9429397583007812,
|
| 455 |
+
"learning_rate": 4.8905608755129964e-05,
|
| 456 |
+
"loss": 4.4937,
|
| 457 |
+
"step": 256
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.02,
|
| 461 |
+
"grad_norm": 1.4009640216827393,
|
| 462 |
+
"learning_rate": 4.888850889192887e-05,
|
| 463 |
+
"loss": 4.3625,
|
| 464 |
+
"step": 260
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.02,
|
| 468 |
+
"grad_norm": 1.7445279359817505,
|
| 469 |
+
"learning_rate": 4.887140902872777e-05,
|
| 470 |
+
"loss": 4.2759,
|
| 471 |
+
"step": 264
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.02,
|
| 475 |
+
"grad_norm": 1.4381403923034668,
|
| 476 |
+
"learning_rate": 4.885430916552668e-05,
|
| 477 |
+
"loss": 4.574,
|
| 478 |
+
"step": 268
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.02,
|
| 482 |
+
"grad_norm": 1.5582979917526245,
|
| 483 |
+
"learning_rate": 4.883720930232558e-05,
|
| 484 |
+
"loss": 4.3831,
|
| 485 |
+
"step": 272
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.02,
|
| 489 |
+
"grad_norm": 1.461166501045227,
|
| 490 |
+
"learning_rate": 4.882010943912449e-05,
|
| 491 |
+
"loss": 4.5049,
|
| 492 |
+
"step": 276
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.02,
|
| 496 |
+
"grad_norm": 1.756549596786499,
|
| 497 |
+
"learning_rate": 4.8803009575923394e-05,
|
| 498 |
+
"loss": 4.1334,
|
| 499 |
+
"step": 280
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.02,
|
| 503 |
+
"grad_norm": 1.9006497859954834,
|
| 504 |
+
"learning_rate": 4.87859097127223e-05,
|
| 505 |
+
"loss": 4.3406,
|
| 506 |
+
"step": 284
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.02,
|
| 510 |
+
"grad_norm": 1.4551759958267212,
|
| 511 |
+
"learning_rate": 4.876880984952121e-05,
|
| 512 |
+
"loss": 4.4688,
|
| 513 |
+
"step": 288
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.02,
|
| 517 |
+
"grad_norm": 1.5335747003555298,
|
| 518 |
+
"learning_rate": 4.875170998632011e-05,
|
| 519 |
+
"loss": 4.2096,
|
| 520 |
+
"step": 292
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.03,
|
| 524 |
+
"grad_norm": 1.4802557229995728,
|
| 525 |
+
"learning_rate": 4.873461012311902e-05,
|
| 526 |
+
"loss": 4.0918,
|
| 527 |
+
"step": 296
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.03,
|
| 531 |
+
"grad_norm": 1.9993939399719238,
|
| 532 |
+
"learning_rate": 4.8717510259917924e-05,
|
| 533 |
+
"loss": 4.3636,
|
| 534 |
+
"step": 300
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.03,
|
| 538 |
+
"grad_norm": 1.706895351409912,
|
| 539 |
+
"learning_rate": 4.8700410396716825e-05,
|
| 540 |
+
"loss": 4.0891,
|
| 541 |
+
"step": 304
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.03,
|
| 545 |
+
"grad_norm": 1.4630484580993652,
|
| 546 |
+
"learning_rate": 4.868331053351574e-05,
|
| 547 |
+
"loss": 4.3254,
|
| 548 |
+
"step": 308
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.03,
|
| 552 |
+
"grad_norm": 1.4353913068771362,
|
| 553 |
+
"learning_rate": 4.866621067031464e-05,
|
| 554 |
+
"loss": 4.1935,
|
| 555 |
+
"step": 312
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.03,
|
| 559 |
+
"grad_norm": 1.5042275190353394,
|
| 560 |
+
"learning_rate": 4.864911080711355e-05,
|
| 561 |
+
"loss": 4.0913,
|
| 562 |
+
"step": 316
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.03,
|
| 566 |
+
"grad_norm": 1.792472004890442,
|
| 567 |
+
"learning_rate": 4.863201094391245e-05,
|
| 568 |
+
"loss": 4.2159,
|
| 569 |
+
"step": 320
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.03,
|
| 573 |
+
"grad_norm": 1.4059948921203613,
|
| 574 |
+
"learning_rate": 4.8614911080711355e-05,
|
| 575 |
+
"loss": 4.2934,
|
| 576 |
+
"step": 324
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.03,
|
| 580 |
+
"grad_norm": 1.8408161401748657,
|
| 581 |
+
"learning_rate": 4.859781121751026e-05,
|
| 582 |
+
"loss": 4.1482,
|
| 583 |
+
"step": 328
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.03,
|
| 587 |
+
"grad_norm": 1.716046690940857,
|
| 588 |
+
"learning_rate": 4.858071135430917e-05,
|
| 589 |
+
"loss": 4.0273,
|
| 590 |
+
"step": 332
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.03,
|
| 594 |
+
"grad_norm": 1.8415429592132568,
|
| 595 |
+
"learning_rate": 4.856361149110808e-05,
|
| 596 |
+
"loss": 4.0837,
|
| 597 |
+
"step": 336
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.03,
|
| 601 |
+
"grad_norm": 1.4880731105804443,
|
| 602 |
+
"learning_rate": 4.854651162790698e-05,
|
| 603 |
+
"loss": 4.1957,
|
| 604 |
+
"step": 340
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.03,
|
| 608 |
+
"grad_norm": 1.8150016069412231,
|
| 609 |
+
"learning_rate": 4.8529411764705885e-05,
|
| 610 |
+
"loss": 4.225,
|
| 611 |
+
"step": 344
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.03,
|
| 615 |
+
"grad_norm": 1.7828190326690674,
|
| 616 |
+
"learning_rate": 4.851231190150479e-05,
|
| 617 |
+
"loss": 4.1053,
|
| 618 |
+
"step": 348
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.03,
|
| 622 |
+
"grad_norm": 1.8549565076828003,
|
| 623 |
+
"learning_rate": 4.849521203830369e-05,
|
| 624 |
+
"loss": 3.8068,
|
| 625 |
+
"step": 352
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.03,
|
| 629 |
+
"grad_norm": 1.4295508861541748,
|
| 630 |
+
"learning_rate": 4.84781121751026e-05,
|
| 631 |
+
"loss": 3.9358,
|
| 632 |
+
"step": 356
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.03,
|
| 636 |
+
"grad_norm": 1.4277971982955933,
|
| 637 |
+
"learning_rate": 4.846101231190151e-05,
|
| 638 |
+
"loss": 4.1512,
|
| 639 |
+
"step": 360
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.03,
|
| 643 |
+
"grad_norm": 1.4937466382980347,
|
| 644 |
+
"learning_rate": 4.8443912448700415e-05,
|
| 645 |
+
"loss": 4.0162,
|
| 646 |
+
"step": 364
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.03,
|
| 650 |
+
"grad_norm": 1.3076913356781006,
|
| 651 |
+
"learning_rate": 4.842681258549932e-05,
|
| 652 |
+
"loss": 4.0336,
|
| 653 |
+
"step": 368
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.03,
|
| 657 |
+
"grad_norm": 1.673946738243103,
|
| 658 |
+
"learning_rate": 4.840971272229822e-05,
|
| 659 |
+
"loss": 4.0915,
|
| 660 |
+
"step": 372
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.03,
|
| 664 |
+
"grad_norm": 1.5978224277496338,
|
| 665 |
+
"learning_rate": 4.839261285909713e-05,
|
| 666 |
+
"loss": 4.137,
|
| 667 |
+
"step": 376
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.03,
|
| 671 |
+
"grad_norm": 2.214573383331299,
|
| 672 |
+
"learning_rate": 4.837551299589603e-05,
|
| 673 |
+
"loss": 3.8231,
|
| 674 |
+
"step": 380
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.03,
|
| 678 |
+
"grad_norm": 1.4263286590576172,
|
| 679 |
+
"learning_rate": 4.835841313269494e-05,
|
| 680 |
+
"loss": 3.9326,
|
| 681 |
+
"step": 384
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.03,
|
| 685 |
+
"grad_norm": 1.694931149482727,
|
| 686 |
+
"learning_rate": 4.8341313269493845e-05,
|
| 687 |
+
"loss": 4.0673,
|
| 688 |
+
"step": 388
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.03,
|
| 692 |
+
"grad_norm": 1.558498740196228,
|
| 693 |
+
"learning_rate": 4.832421340629275e-05,
|
| 694 |
+
"loss": 3.9454,
|
| 695 |
+
"step": 392
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.03,
|
| 699 |
+
"grad_norm": 1.5231866836547852,
|
| 700 |
+
"learning_rate": 4.830711354309166e-05,
|
| 701 |
+
"loss": 4.0214,
|
| 702 |
+
"step": 396
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.03,
|
| 706 |
+
"grad_norm": 1.6748324632644653,
|
| 707 |
+
"learning_rate": 4.829001367989056e-05,
|
| 708 |
+
"loss": 3.9983,
|
| 709 |
+
"step": 400
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.03,
|
| 713 |
+
"grad_norm": 1.3733506202697754,
|
| 714 |
+
"learning_rate": 4.827291381668947e-05,
|
| 715 |
+
"loss": 3.8526,
|
| 716 |
+
"step": 404
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.03,
|
| 720 |
+
"grad_norm": 1.6695655584335327,
|
| 721 |
+
"learning_rate": 4.8255813953488375e-05,
|
| 722 |
+
"loss": 3.8948,
|
| 723 |
+
"step": 408
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.04,
|
| 727 |
+
"grad_norm": 1.3915668725967407,
|
| 728 |
+
"learning_rate": 4.8238714090287276e-05,
|
| 729 |
+
"loss": 3.8511,
|
| 730 |
+
"step": 412
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.04,
|
| 734 |
+
"grad_norm": 1.5736514329910278,
|
| 735 |
+
"learning_rate": 4.822161422708619e-05,
|
| 736 |
+
"loss": 3.9446,
|
| 737 |
+
"step": 416
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.04,
|
| 741 |
+
"grad_norm": 1.562687635421753,
|
| 742 |
+
"learning_rate": 4.820451436388509e-05,
|
| 743 |
+
"loss": 3.6682,
|
| 744 |
+
"step": 420
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.04,
|
| 748 |
+
"grad_norm": 1.4215232133865356,
|
| 749 |
+
"learning_rate": 4.8187414500684e-05,
|
| 750 |
+
"loss": 3.6618,
|
| 751 |
+
"step": 424
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.04,
|
| 755 |
+
"grad_norm": 1.8750267028808594,
|
| 756 |
+
"learning_rate": 4.8170314637482905e-05,
|
| 757 |
+
"loss": 3.8246,
|
| 758 |
+
"step": 428
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.04,
|
| 762 |
+
"grad_norm": 1.5843207836151123,
|
| 763 |
+
"learning_rate": 4.8153214774281806e-05,
|
| 764 |
+
"loss": 3.807,
|
| 765 |
+
"step": 432
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.04,
|
| 769 |
+
"grad_norm": 1.6384755373001099,
|
| 770 |
+
"learning_rate": 4.813611491108071e-05,
|
| 771 |
+
"loss": 3.8329,
|
| 772 |
+
"step": 436
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.04,
|
| 776 |
+
"grad_norm": 1.646612286567688,
|
| 777 |
+
"learning_rate": 4.811901504787962e-05,
|
| 778 |
+
"loss": 3.9736,
|
| 779 |
+
"step": 440
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.04,
|
| 783 |
+
"grad_norm": 1.592463493347168,
|
| 784 |
+
"learning_rate": 4.810191518467853e-05,
|
| 785 |
+
"loss": 3.6107,
|
| 786 |
+
"step": 444
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.04,
|
| 790 |
+
"grad_norm": 1.6230803728103638,
|
| 791 |
+
"learning_rate": 4.808481532147743e-05,
|
| 792 |
+
"loss": 3.8141,
|
| 793 |
+
"step": 448
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.04,
|
| 797 |
+
"grad_norm": 1.7098625898361206,
|
| 798 |
+
"learning_rate": 4.8067715458276336e-05,
|
| 799 |
+
"loss": 3.8751,
|
| 800 |
+
"step": 452
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.04,
|
| 804 |
+
"grad_norm": 1.441146731376648,
|
| 805 |
+
"learning_rate": 4.805061559507524e-05,
|
| 806 |
+
"loss": 3.8461,
|
| 807 |
+
"step": 456
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.04,
|
| 811 |
+
"grad_norm": 1.4387036561965942,
|
| 812 |
+
"learning_rate": 4.8033515731874144e-05,
|
| 813 |
+
"loss": 3.8432,
|
| 814 |
+
"step": 460
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.04,
|
| 818 |
+
"grad_norm": 1.6620376110076904,
|
| 819 |
+
"learning_rate": 4.801641586867305e-05,
|
| 820 |
+
"loss": 3.7844,
|
| 821 |
+
"step": 464
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.04,
|
| 825 |
+
"grad_norm": 1.5403114557266235,
|
| 826 |
+
"learning_rate": 4.799931600547196e-05,
|
| 827 |
+
"loss": 3.6556,
|
| 828 |
+
"step": 468
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.04,
|
| 832 |
+
"grad_norm": 1.5642478466033936,
|
| 833 |
+
"learning_rate": 4.7982216142270866e-05,
|
| 834 |
+
"loss": 3.7526,
|
| 835 |
+
"step": 472
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.04,
|
| 839 |
+
"grad_norm": 1.5027506351470947,
|
| 840 |
+
"learning_rate": 4.796511627906977e-05,
|
| 841 |
+
"loss": 3.7862,
|
| 842 |
+
"step": 476
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.04,
|
| 846 |
+
"grad_norm": 1.5570485591888428,
|
| 847 |
+
"learning_rate": 4.7948016415868674e-05,
|
| 848 |
+
"loss": 3.7656,
|
| 849 |
+
"step": 480
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.04,
|
| 853 |
+
"grad_norm": 2.2708029747009277,
|
| 854 |
+
"learning_rate": 4.793091655266758e-05,
|
| 855 |
+
"loss": 3.8742,
|
| 856 |
+
"step": 484
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.04,
|
| 860 |
+
"grad_norm": 1.5446516275405884,
|
| 861 |
+
"learning_rate": 4.791381668946648e-05,
|
| 862 |
+
"loss": 3.6401,
|
| 863 |
+
"step": 488
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.04,
|
| 867 |
+
"grad_norm": 1.5480107069015503,
|
| 868 |
+
"learning_rate": 4.789671682626539e-05,
|
| 869 |
+
"loss": 3.7441,
|
| 870 |
+
"step": 492
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.04,
|
| 874 |
+
"grad_norm": 1.5571659803390503,
|
| 875 |
+
"learning_rate": 4.78796169630643e-05,
|
| 876 |
+
"loss": 3.6849,
|
| 877 |
+
"step": 496
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.04,
|
| 881 |
+
"grad_norm": 1.9021155834197998,
|
| 882 |
+
"learning_rate": 4.7862517099863204e-05,
|
| 883 |
+
"loss": 3.7989,
|
| 884 |
+
"step": 500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.04,
|
| 888 |
+
"grad_norm": 1.5292832851409912,
|
| 889 |
+
"learning_rate": 4.784541723666211e-05,
|
| 890 |
+
"loss": 3.8129,
|
| 891 |
+
"step": 504
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.04,
|
| 895 |
+
"grad_norm": 1.5750986337661743,
|
| 896 |
+
"learning_rate": 4.782831737346101e-05,
|
| 897 |
+
"loss": 3.648,
|
| 898 |
+
"step": 508
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.04,
|
| 902 |
+
"grad_norm": 1.6995611190795898,
|
| 903 |
+
"learning_rate": 4.781121751025992e-05,
|
| 904 |
+
"loss": 3.7686,
|
| 905 |
+
"step": 512
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.04,
|
| 909 |
+
"grad_norm": 1.7573764324188232,
|
| 910 |
+
"learning_rate": 4.7794117647058826e-05,
|
| 911 |
+
"loss": 3.6636,
|
| 912 |
+
"step": 516
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.04,
|
| 916 |
+
"grad_norm": 1.5131741762161255,
|
| 917 |
+
"learning_rate": 4.7777017783857733e-05,
|
| 918 |
+
"loss": 3.5793,
|
| 919 |
+
"step": 520
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.04,
|
| 923 |
+
"grad_norm": 1.6838548183441162,
|
| 924 |
+
"learning_rate": 4.775991792065664e-05,
|
| 925 |
+
"loss": 3.6336,
|
| 926 |
+
"step": 524
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.05,
|
| 930 |
+
"grad_norm": 1.9402954578399658,
|
| 931 |
+
"learning_rate": 4.774281805745554e-05,
|
| 932 |
+
"loss": 3.6733,
|
| 933 |
+
"step": 528
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.05,
|
| 937 |
+
"grad_norm": 1.7752444744110107,
|
| 938 |
+
"learning_rate": 4.772571819425445e-05,
|
| 939 |
+
"loss": 3.8021,
|
| 940 |
+
"step": 532
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.05,
|
| 944 |
+
"grad_norm": 1.4470579624176025,
|
| 945 |
+
"learning_rate": 4.7708618331053356e-05,
|
| 946 |
+
"loss": 3.6526,
|
| 947 |
+
"step": 536
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.05,
|
| 951 |
+
"grad_norm": 1.5560115575790405,
|
| 952 |
+
"learning_rate": 4.769151846785226e-05,
|
| 953 |
+
"loss": 3.7236,
|
| 954 |
+
"step": 540
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.05,
|
| 958 |
+
"grad_norm": 1.6849740743637085,
|
| 959 |
+
"learning_rate": 4.7674418604651164e-05,
|
| 960 |
+
"loss": 3.7586,
|
| 961 |
+
"step": 544
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.05,
|
| 965 |
+
"grad_norm": 1.4917711019515991,
|
| 966 |
+
"learning_rate": 4.765731874145007e-05,
|
| 967 |
+
"loss": 3.7791,
|
| 968 |
+
"step": 548
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.05,
|
| 972 |
+
"grad_norm": 1.791801929473877,
|
| 973 |
+
"learning_rate": 4.764021887824898e-05,
|
| 974 |
+
"loss": 3.5659,
|
| 975 |
+
"step": 552
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.05,
|
| 979 |
+
"grad_norm": 1.531817078590393,
|
| 980 |
+
"learning_rate": 4.7623119015047886e-05,
|
| 981 |
+
"loss": 3.5811,
|
| 982 |
+
"step": 556
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.05,
|
| 986 |
+
"grad_norm": 1.8725793361663818,
|
| 987 |
+
"learning_rate": 4.7606019151846787e-05,
|
| 988 |
+
"loss": 3.7337,
|
| 989 |
+
"step": 560
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.05,
|
| 993 |
+
"grad_norm": 1.5851105451583862,
|
| 994 |
+
"learning_rate": 4.7588919288645694e-05,
|
| 995 |
+
"loss": 3.4899,
|
| 996 |
+
"step": 564
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.05,
|
| 1000 |
+
"grad_norm": 1.4937329292297363,
|
| 1001 |
+
"learning_rate": 4.7571819425444594e-05,
|
| 1002 |
+
"loss": 3.5653,
|
| 1003 |
+
"step": 568
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.05,
|
| 1007 |
+
"grad_norm": 1.8102850914001465,
|
| 1008 |
+
"learning_rate": 4.75547195622435e-05,
|
| 1009 |
+
"loss": 3.517,
|
| 1010 |
+
"step": 572
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.05,
|
| 1014 |
+
"grad_norm": 1.4572982788085938,
|
| 1015 |
+
"learning_rate": 4.753761969904241e-05,
|
| 1016 |
+
"loss": 3.723,
|
| 1017 |
+
"step": 576
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.05,
|
| 1021 |
+
"grad_norm": 1.5815645456314087,
|
| 1022 |
+
"learning_rate": 4.7520519835841317e-05,
|
| 1023 |
+
"loss": 3.6615,
|
| 1024 |
+
"step": 580
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.05,
|
| 1028 |
+
"grad_norm": 1.6406457424163818,
|
| 1029 |
+
"learning_rate": 4.7503419972640224e-05,
|
| 1030 |
+
"loss": 3.6796,
|
| 1031 |
+
"step": 584
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.05,
|
| 1035 |
+
"grad_norm": 1.8512486219406128,
|
| 1036 |
+
"learning_rate": 4.7486320109439124e-05,
|
| 1037 |
+
"loss": 3.5586,
|
| 1038 |
+
"step": 588
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.05,
|
| 1042 |
+
"grad_norm": 1.6738507747650146,
|
| 1043 |
+
"learning_rate": 4.746922024623803e-05,
|
| 1044 |
+
"loss": 3.4724,
|
| 1045 |
+
"step": 592
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.05,
|
| 1049 |
+
"grad_norm": 1.766518235206604,
|
| 1050 |
+
"learning_rate": 4.745212038303693e-05,
|
| 1051 |
+
"loss": 3.7213,
|
| 1052 |
+
"step": 596
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.05,
|
| 1056 |
+
"grad_norm": 1.681229591369629,
|
| 1057 |
+
"learning_rate": 4.7435020519835846e-05,
|
| 1058 |
+
"loss": 3.6538,
|
| 1059 |
+
"step": 600
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.05,
|
| 1063 |
+
"grad_norm": 1.7900785207748413,
|
| 1064 |
+
"learning_rate": 4.7417920656634754e-05,
|
| 1065 |
+
"loss": 3.7952,
|
| 1066 |
+
"step": 604
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.05,
|
| 1070 |
+
"grad_norm": 1.5610637664794922,
|
| 1071 |
+
"learning_rate": 4.7400820793433654e-05,
|
| 1072 |
+
"loss": 3.5255,
|
| 1073 |
+
"step": 608
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.05,
|
| 1077 |
+
"grad_norm": 1.3365826606750488,
|
| 1078 |
+
"learning_rate": 4.738372093023256e-05,
|
| 1079 |
+
"loss": 3.5621,
|
| 1080 |
+
"step": 612
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.05,
|
| 1084 |
+
"grad_norm": 1.8946609497070312,
|
| 1085 |
+
"learning_rate": 4.736662106703146e-05,
|
| 1086 |
+
"loss": 3.5664,
|
| 1087 |
+
"step": 616
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.05,
|
| 1091 |
+
"grad_norm": 1.728330135345459,
|
| 1092 |
+
"learning_rate": 4.734952120383037e-05,
|
| 1093 |
+
"loss": 3.7577,
|
| 1094 |
+
"step": 620
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.05,
|
| 1098 |
+
"grad_norm": 1.7583847045898438,
|
| 1099 |
+
"learning_rate": 4.733242134062928e-05,
|
| 1100 |
+
"loss": 3.5722,
|
| 1101 |
+
"step": 624
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.05,
|
| 1105 |
+
"grad_norm": 1.9061710834503174,
|
| 1106 |
+
"learning_rate": 4.7315321477428184e-05,
|
| 1107 |
+
"loss": 3.3436,
|
| 1108 |
+
"step": 628
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.05,
|
| 1112 |
+
"grad_norm": 1.7074024677276611,
|
| 1113 |
+
"learning_rate": 4.729822161422709e-05,
|
| 1114 |
+
"loss": 3.5426,
|
| 1115 |
+
"step": 632
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.05,
|
| 1119 |
+
"grad_norm": 1.649793028831482,
|
| 1120 |
+
"learning_rate": 4.728112175102599e-05,
|
| 1121 |
+
"loss": 3.5826,
|
| 1122 |
+
"step": 636
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.05,
|
| 1126 |
+
"grad_norm": 1.7260462045669556,
|
| 1127 |
+
"learning_rate": 4.72640218878249e-05,
|
| 1128 |
+
"loss": 3.5353,
|
| 1129 |
+
"step": 640
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.06,
|
| 1133 |
+
"grad_norm": 1.745391607284546,
|
| 1134 |
+
"learning_rate": 4.724692202462381e-05,
|
| 1135 |
+
"loss": 3.3534,
|
| 1136 |
+
"step": 644
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.06,
|
| 1140 |
+
"grad_norm": 1.374226450920105,
|
| 1141 |
+
"learning_rate": 4.722982216142271e-05,
|
| 1142 |
+
"loss": 3.386,
|
| 1143 |
+
"step": 648
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.06,
|
| 1147 |
+
"grad_norm": 1.8732798099517822,
|
| 1148 |
+
"learning_rate": 4.7212722298221615e-05,
|
| 1149 |
+
"loss": 3.6073,
|
| 1150 |
+
"step": 652
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.06,
|
| 1154 |
+
"grad_norm": 2.4075376987457275,
|
| 1155 |
+
"learning_rate": 4.719562243502052e-05,
|
| 1156 |
+
"loss": 3.5393,
|
| 1157 |
+
"step": 656
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.06,
|
| 1161 |
+
"grad_norm": 1.54508638381958,
|
| 1162 |
+
"learning_rate": 4.717852257181943e-05,
|
| 1163 |
+
"loss": 3.475,
|
| 1164 |
+
"step": 660
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.06,
|
| 1168 |
+
"grad_norm": 1.5737498998641968,
|
| 1169 |
+
"learning_rate": 4.716142270861834e-05,
|
| 1170 |
+
"loss": 3.5497,
|
| 1171 |
+
"step": 664
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.06,
|
| 1175 |
+
"grad_norm": 1.8623074293136597,
|
| 1176 |
+
"learning_rate": 4.714432284541724e-05,
|
| 1177 |
+
"loss": 3.4603,
|
| 1178 |
+
"step": 668
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.06,
|
| 1182 |
+
"grad_norm": 1.7199251651763916,
|
| 1183 |
+
"learning_rate": 4.7127222982216145e-05,
|
| 1184 |
+
"loss": 3.5801,
|
| 1185 |
+
"step": 672
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.06,
|
| 1189 |
+
"grad_norm": 1.5866843461990356,
|
| 1190 |
+
"learning_rate": 4.7110123119015045e-05,
|
| 1191 |
+
"loss": 3.3454,
|
| 1192 |
+
"step": 676
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.06,
|
| 1196 |
+
"grad_norm": 1.9907779693603516,
|
| 1197 |
+
"learning_rate": 4.709302325581396e-05,
|
| 1198 |
+
"loss": 3.6247,
|
| 1199 |
+
"step": 680
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.06,
|
| 1203 |
+
"grad_norm": 1.3720687627792358,
|
| 1204 |
+
"learning_rate": 4.707592339261287e-05,
|
| 1205 |
+
"loss": 3.3614,
|
| 1206 |
+
"step": 684
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.06,
|
| 1210 |
+
"grad_norm": 1.739660620689392,
|
| 1211 |
+
"learning_rate": 4.705882352941177e-05,
|
| 1212 |
+
"loss": 3.4211,
|
| 1213 |
+
"step": 688
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.06,
|
| 1217 |
+
"grad_norm": 1.6425236463546753,
|
| 1218 |
+
"learning_rate": 4.7041723666210675e-05,
|
| 1219 |
+
"loss": 3.423,
|
| 1220 |
+
"step": 692
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.06,
|
| 1224 |
+
"grad_norm": 1.5457091331481934,
|
| 1225 |
+
"learning_rate": 4.7024623803009575e-05,
|
| 1226 |
+
"loss": 3.3258,
|
| 1227 |
+
"step": 696
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.06,
|
| 1231 |
+
"grad_norm": 1.5979949235916138,
|
| 1232 |
+
"learning_rate": 4.700752393980848e-05,
|
| 1233 |
+
"loss": 3.2838,
|
| 1234 |
+
"step": 700
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.06,
|
| 1238 |
+
"grad_norm": 1.6761040687561035,
|
| 1239 |
+
"learning_rate": 4.699042407660739e-05,
|
| 1240 |
+
"loss": 3.334,
|
| 1241 |
+
"step": 704
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.06,
|
| 1245 |
+
"grad_norm": 1.552573323249817,
|
| 1246 |
+
"learning_rate": 4.69733242134063e-05,
|
| 1247 |
+
"loss": 3.5355,
|
| 1248 |
+
"step": 708
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.06,
|
| 1252 |
+
"grad_norm": 1.6743354797363281,
|
| 1253 |
+
"learning_rate": 4.6956224350205205e-05,
|
| 1254 |
+
"loss": 3.4076,
|
| 1255 |
+
"step": 712
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.06,
|
| 1259 |
+
"grad_norm": 4.555662155151367,
|
| 1260 |
+
"learning_rate": 4.6939124487004105e-05,
|
| 1261 |
+
"loss": 3.7446,
|
| 1262 |
+
"step": 716
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.06,
|
| 1266 |
+
"grad_norm": 1.5942860841751099,
|
| 1267 |
+
"learning_rate": 4.692202462380301e-05,
|
| 1268 |
+
"loss": 3.553,
|
| 1269 |
+
"step": 720
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.06,
|
| 1273 |
+
"grad_norm": 1.9089189767837524,
|
| 1274 |
+
"learning_rate": 4.690492476060191e-05,
|
| 1275 |
+
"loss": 3.3584,
|
| 1276 |
+
"step": 724
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.06,
|
| 1280 |
+
"grad_norm": 2.0768373012542725,
|
| 1281 |
+
"learning_rate": 4.688782489740082e-05,
|
| 1282 |
+
"loss": 3.4295,
|
| 1283 |
+
"step": 728
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.06,
|
| 1287 |
+
"grad_norm": 3.3060362339019775,
|
| 1288 |
+
"learning_rate": 4.687072503419973e-05,
|
| 1289 |
+
"loss": 3.415,
|
| 1290 |
+
"step": 732
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.06,
|
| 1294 |
+
"grad_norm": 1.9307421445846558,
|
| 1295 |
+
"learning_rate": 4.6853625170998635e-05,
|
| 1296 |
+
"loss": 3.4901,
|
| 1297 |
+
"step": 736
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.06,
|
| 1301 |
+
"grad_norm": 1.7142446041107178,
|
| 1302 |
+
"learning_rate": 4.683652530779754e-05,
|
| 1303 |
+
"loss": 3.6096,
|
| 1304 |
+
"step": 740
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.06,
|
| 1308 |
+
"grad_norm": 1.7756657600402832,
|
| 1309 |
+
"learning_rate": 4.681942544459644e-05,
|
| 1310 |
+
"loss": 3.4534,
|
| 1311 |
+
"step": 744
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.06,
|
| 1315 |
+
"grad_norm": 1.5459442138671875,
|
| 1316 |
+
"learning_rate": 4.680232558139535e-05,
|
| 1317 |
+
"loss": 3.3813,
|
| 1318 |
+
"step": 748
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.06,
|
| 1322 |
+
"grad_norm": 2.1961963176727295,
|
| 1323 |
+
"learning_rate": 4.678522571819426e-05,
|
| 1324 |
+
"loss": 3.3467,
|
| 1325 |
+
"step": 752
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.06,
|
| 1329 |
+
"grad_norm": 1.7320613861083984,
|
| 1330 |
+
"learning_rate": 4.676812585499316e-05,
|
| 1331 |
+
"loss": 3.3927,
|
| 1332 |
+
"step": 756
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.06,
|
| 1336 |
+
"grad_norm": 1.6255303621292114,
|
| 1337 |
+
"learning_rate": 4.6751025991792066e-05,
|
| 1338 |
+
"loss": 3.4644,
|
| 1339 |
+
"step": 760
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.07,
|
| 1343 |
+
"grad_norm": 1.5142388343811035,
|
| 1344 |
+
"learning_rate": 4.673392612859097e-05,
|
| 1345 |
+
"loss": 3.4657,
|
| 1346 |
+
"step": 764
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.07,
|
| 1350 |
+
"grad_norm": 1.5093744993209839,
|
| 1351 |
+
"learning_rate": 4.671682626538988e-05,
|
| 1352 |
+
"loss": 3.3144,
|
| 1353 |
+
"step": 768
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.07,
|
| 1357 |
+
"grad_norm": 1.4730446338653564,
|
| 1358 |
+
"learning_rate": 4.669972640218879e-05,
|
| 1359 |
+
"loss": 3.3382,
|
| 1360 |
+
"step": 772
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.07,
|
| 1364 |
+
"grad_norm": 1.7650116682052612,
|
| 1365 |
+
"learning_rate": 4.668262653898769e-05,
|
| 1366 |
+
"loss": 3.4519,
|
| 1367 |
+
"step": 776
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.07,
|
| 1371 |
+
"grad_norm": 1.637071132659912,
|
| 1372 |
+
"learning_rate": 4.6665526675786596e-05,
|
| 1373 |
+
"loss": 3.3723,
|
| 1374 |
+
"step": 780
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.07,
|
| 1378 |
+
"grad_norm": 1.4582788944244385,
|
| 1379 |
+
"learning_rate": 4.6648426812585496e-05,
|
| 1380 |
+
"loss": 3.4839,
|
| 1381 |
+
"step": 784
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.07,
|
| 1385 |
+
"grad_norm": 1.4654922485351562,
|
| 1386 |
+
"learning_rate": 4.663132694938441e-05,
|
| 1387 |
+
"loss": 3.1257,
|
| 1388 |
+
"step": 788
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.07,
|
| 1392 |
+
"grad_norm": 1.4668978452682495,
|
| 1393 |
+
"learning_rate": 4.661422708618332e-05,
|
| 1394 |
+
"loss": 3.2506,
|
| 1395 |
+
"step": 792
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.07,
|
| 1399 |
+
"grad_norm": 1.5190536975860596,
|
| 1400 |
+
"learning_rate": 4.659712722298222e-05,
|
| 1401 |
+
"loss": 3.4055,
|
| 1402 |
+
"step": 796
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.07,
|
| 1406 |
+
"grad_norm": 1.9057687520980835,
|
| 1407 |
+
"learning_rate": 4.6580027359781126e-05,
|
| 1408 |
+
"loss": 3.4804,
|
| 1409 |
+
"step": 800
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.07,
|
| 1413 |
+
"grad_norm": 1.4573813676834106,
|
| 1414 |
+
"learning_rate": 4.6562927496580026e-05,
|
| 1415 |
+
"loss": 3.3792,
|
| 1416 |
+
"step": 804
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.07,
|
| 1420 |
+
"grad_norm": 2.3757171630859375,
|
| 1421 |
+
"learning_rate": 4.6545827633378933e-05,
|
| 1422 |
+
"loss": 3.4649,
|
| 1423 |
+
"step": 808
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.07,
|
| 1427 |
+
"grad_norm": 1.7024015188217163,
|
| 1428 |
+
"learning_rate": 4.652872777017784e-05,
|
| 1429 |
+
"loss": 3.3149,
|
| 1430 |
+
"step": 812
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.07,
|
| 1434 |
+
"grad_norm": 1.5424153804779053,
|
| 1435 |
+
"learning_rate": 4.651162790697675e-05,
|
| 1436 |
+
"loss": 3.3,
|
| 1437 |
+
"step": 816
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.07,
|
| 1441 |
+
"grad_norm": 1.4667370319366455,
|
| 1442 |
+
"learning_rate": 4.6494528043775655e-05,
|
| 1443 |
+
"loss": 3.2845,
|
| 1444 |
+
"step": 820
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.07,
|
| 1448 |
+
"grad_norm": 1.5216193199157715,
|
| 1449 |
+
"learning_rate": 4.6477428180574556e-05,
|
| 1450 |
+
"loss": 3.3062,
|
| 1451 |
+
"step": 824
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.07,
|
| 1455 |
+
"grad_norm": 1.5499827861785889,
|
| 1456 |
+
"learning_rate": 4.6460328317373463e-05,
|
| 1457 |
+
"loss": 3.263,
|
| 1458 |
+
"step": 828
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.07,
|
| 1462 |
+
"grad_norm": 1.644148588180542,
|
| 1463 |
+
"learning_rate": 4.6443228454172364e-05,
|
| 1464 |
+
"loss": 3.4374,
|
| 1465 |
+
"step": 832
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.07,
|
| 1469 |
+
"grad_norm": 1.6420782804489136,
|
| 1470 |
+
"learning_rate": 4.642612859097127e-05,
|
| 1471 |
+
"loss": 3.4049,
|
| 1472 |
+
"step": 836
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.07,
|
| 1476 |
+
"grad_norm": 1.4325385093688965,
|
| 1477 |
+
"learning_rate": 4.640902872777018e-05,
|
| 1478 |
+
"loss": 3.3499,
|
| 1479 |
+
"step": 840
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.07,
|
| 1483 |
+
"grad_norm": 1.6447879076004028,
|
| 1484 |
+
"learning_rate": 4.6391928864569086e-05,
|
| 1485 |
+
"loss": 3.4098,
|
| 1486 |
+
"step": 844
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.07,
|
| 1490 |
+
"grad_norm": 1.4675137996673584,
|
| 1491 |
+
"learning_rate": 4.637482900136799e-05,
|
| 1492 |
+
"loss": 3.2778,
|
| 1493 |
+
"step": 848
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.07,
|
| 1497 |
+
"grad_norm": 1.5847750902175903,
|
| 1498 |
+
"learning_rate": 4.6357729138166894e-05,
|
| 1499 |
+
"loss": 3.1695,
|
| 1500 |
+
"step": 852
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.07,
|
| 1504 |
+
"grad_norm": 1.514772653579712,
|
| 1505 |
+
"learning_rate": 4.63406292749658e-05,
|
| 1506 |
+
"loss": 3.1556,
|
| 1507 |
+
"step": 856
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.07,
|
| 1511 |
+
"grad_norm": 1.511367917060852,
|
| 1512 |
+
"learning_rate": 4.632352941176471e-05,
|
| 1513 |
+
"loss": 3.3307,
|
| 1514 |
+
"step": 860
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.07,
|
| 1518 |
+
"grad_norm": 1.7640196084976196,
|
| 1519 |
+
"learning_rate": 4.630642954856361e-05,
|
| 1520 |
+
"loss": 3.2985,
|
| 1521 |
+
"step": 864
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.07,
|
| 1525 |
+
"grad_norm": 1.7074265480041504,
|
| 1526 |
+
"learning_rate": 4.628932968536252e-05,
|
| 1527 |
+
"loss": 3.2598,
|
| 1528 |
+
"step": 868
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.07,
|
| 1532 |
+
"grad_norm": 1.52577805519104,
|
| 1533 |
+
"learning_rate": 4.6272229822161424e-05,
|
| 1534 |
+
"loss": 3.0872,
|
| 1535 |
+
"step": 872
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.07,
|
| 1539 |
+
"grad_norm": 1.6908501386642456,
|
| 1540 |
+
"learning_rate": 4.625512995896033e-05,
|
| 1541 |
+
"loss": 3.1931,
|
| 1542 |
+
"step": 876
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.08,
|
| 1546 |
+
"grad_norm": 1.6583012342453003,
|
| 1547 |
+
"learning_rate": 4.623803009575924e-05,
|
| 1548 |
+
"loss": 3.1849,
|
| 1549 |
+
"step": 880
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.08,
|
| 1553 |
+
"grad_norm": 1.4067027568817139,
|
| 1554 |
+
"learning_rate": 4.622093023255814e-05,
|
| 1555 |
+
"loss": 3.1912,
|
| 1556 |
+
"step": 884
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.08,
|
| 1560 |
+
"grad_norm": 1.388487458229065,
|
| 1561 |
+
"learning_rate": 4.6203830369357046e-05,
|
| 1562 |
+
"loss": 3.2845,
|
| 1563 |
+
"step": 888
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.08,
|
| 1567 |
+
"grad_norm": 1.7910948991775513,
|
| 1568 |
+
"learning_rate": 4.6186730506155954e-05,
|
| 1569 |
+
"loss": 3.1146,
|
| 1570 |
+
"step": 892
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.08,
|
| 1574 |
+
"grad_norm": 1.4454721212387085,
|
| 1575 |
+
"learning_rate": 4.616963064295486e-05,
|
| 1576 |
+
"loss": 3.2114,
|
| 1577 |
+
"step": 896
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.08,
|
| 1581 |
+
"grad_norm": 1.4809476137161255,
|
| 1582 |
+
"learning_rate": 4.615253077975377e-05,
|
| 1583 |
+
"loss": 3.3329,
|
| 1584 |
+
"step": 900
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.08,
|
| 1588 |
+
"grad_norm": 1.6668082475662231,
|
| 1589 |
+
"learning_rate": 4.613543091655267e-05,
|
| 1590 |
+
"loss": 3.2517,
|
| 1591 |
+
"step": 904
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.08,
|
| 1595 |
+
"grad_norm": 1.72063410282135,
|
| 1596 |
+
"learning_rate": 4.6118331053351576e-05,
|
| 1597 |
+
"loss": 3.0895,
|
| 1598 |
+
"step": 908
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.08,
|
| 1602 |
+
"grad_norm": 1.5995526313781738,
|
| 1603 |
+
"learning_rate": 4.610123119015048e-05,
|
| 1604 |
+
"loss": 3.2703,
|
| 1605 |
+
"step": 912
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.08,
|
| 1609 |
+
"grad_norm": 1.453747034072876,
|
| 1610 |
+
"learning_rate": 4.6084131326949384e-05,
|
| 1611 |
+
"loss": 3.3619,
|
| 1612 |
+
"step": 916
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.08,
|
| 1616 |
+
"grad_norm": 1.7982878684997559,
|
| 1617 |
+
"learning_rate": 4.606703146374829e-05,
|
| 1618 |
+
"loss": 3.2657,
|
| 1619 |
+
"step": 920
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.08,
|
| 1623 |
+
"grad_norm": 1.3937926292419434,
|
| 1624 |
+
"learning_rate": 4.60499316005472e-05,
|
| 1625 |
+
"loss": 3.1727,
|
| 1626 |
+
"step": 924
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.08,
|
| 1630 |
+
"grad_norm": 1.5705368518829346,
|
| 1631 |
+
"learning_rate": 4.6032831737346106e-05,
|
| 1632 |
+
"loss": 3.0657,
|
| 1633 |
+
"step": 928
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.08,
|
| 1637 |
+
"grad_norm": 1.6821500062942505,
|
| 1638 |
+
"learning_rate": 4.601573187414501e-05,
|
| 1639 |
+
"loss": 3.3191,
|
| 1640 |
+
"step": 932
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.08,
|
| 1644 |
+
"grad_norm": 1.5288244485855103,
|
| 1645 |
+
"learning_rate": 4.5998632010943914e-05,
|
| 1646 |
+
"loss": 3.0876,
|
| 1647 |
+
"step": 936
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.08,
|
| 1651 |
+
"grad_norm": 1.7748781442642212,
|
| 1652 |
+
"learning_rate": 4.598153214774282e-05,
|
| 1653 |
+
"loss": 3.18,
|
| 1654 |
+
"step": 940
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.08,
|
| 1658 |
+
"grad_norm": 1.7024669647216797,
|
| 1659 |
+
"learning_rate": 4.596443228454172e-05,
|
| 1660 |
+
"loss": 3.3601,
|
| 1661 |
+
"step": 944
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.08,
|
| 1665 |
+
"grad_norm": 1.7017521858215332,
|
| 1666 |
+
"learning_rate": 4.5947332421340636e-05,
|
| 1667 |
+
"loss": 2.9751,
|
| 1668 |
+
"step": 948
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.08,
|
| 1672 |
+
"grad_norm": 1.7405025959014893,
|
| 1673 |
+
"learning_rate": 4.593023255813954e-05,
|
| 1674 |
+
"loss": 3.2073,
|
| 1675 |
+
"step": 952
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.08,
|
| 1679 |
+
"grad_norm": 1.6610547304153442,
|
| 1680 |
+
"learning_rate": 4.5913132694938444e-05,
|
| 1681 |
+
"loss": 3.2363,
|
| 1682 |
+
"step": 956
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.08,
|
| 1686 |
+
"grad_norm": 1.4785616397857666,
|
| 1687 |
+
"learning_rate": 4.5896032831737345e-05,
|
| 1688 |
+
"loss": 3.3254,
|
| 1689 |
+
"step": 960
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.08,
|
| 1693 |
+
"grad_norm": 3.453533411026001,
|
| 1694 |
+
"learning_rate": 4.587893296853625e-05,
|
| 1695 |
+
"loss": 3.21,
|
| 1696 |
+
"step": 964
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.08,
|
| 1700 |
+
"grad_norm": 1.4303959608078003,
|
| 1701 |
+
"learning_rate": 4.586183310533516e-05,
|
| 1702 |
+
"loss": 3.3226,
|
| 1703 |
+
"step": 968
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.08,
|
| 1707 |
+
"grad_norm": 1.6272943019866943,
|
| 1708 |
+
"learning_rate": 4.584473324213407e-05,
|
| 1709 |
+
"loss": 3.5944,
|
| 1710 |
+
"step": 972
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.08,
|
| 1714 |
+
"grad_norm": 1.823716402053833,
|
| 1715 |
+
"learning_rate": 4.5827633378932974e-05,
|
| 1716 |
+
"loss": 3.234,
|
| 1717 |
+
"step": 976
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.08,
|
| 1721 |
+
"grad_norm": 1.7006231546401978,
|
| 1722 |
+
"learning_rate": 4.5810533515731875e-05,
|
| 1723 |
+
"loss": 3.0123,
|
| 1724 |
+
"step": 980
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.08,
|
| 1728 |
+
"grad_norm": 1.3645117282867432,
|
| 1729 |
+
"learning_rate": 4.579343365253078e-05,
|
| 1730 |
+
"loss": 3.2243,
|
| 1731 |
+
"step": 984
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.08,
|
| 1735 |
+
"grad_norm": 1.7724872827529907,
|
| 1736 |
+
"learning_rate": 4.577633378932969e-05,
|
| 1737 |
+
"loss": 3.1705,
|
| 1738 |
+
"step": 988
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.08,
|
| 1742 |
+
"grad_norm": 1.6377599239349365,
|
| 1743 |
+
"learning_rate": 4.575923392612859e-05,
|
| 1744 |
+
"loss": 3.2647,
|
| 1745 |
+
"step": 992
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.09,
|
| 1749 |
+
"grad_norm": 1.4828346967697144,
|
| 1750 |
+
"learning_rate": 4.57421340629275e-05,
|
| 1751 |
+
"loss": 3.0613,
|
| 1752 |
+
"step": 996
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.09,
|
| 1756 |
+
"grad_norm": 1.515553593635559,
|
| 1757 |
+
"learning_rate": 4.5725034199726405e-05,
|
| 1758 |
+
"loss": 3.0368,
|
| 1759 |
+
"step": 1000
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.09,
|
| 1763 |
+
"grad_norm": 1.498844861984253,
|
| 1764 |
+
"learning_rate": 4.570793433652531e-05,
|
| 1765 |
+
"loss": 3.102,
|
| 1766 |
+
"step": 1004
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.09,
|
| 1770 |
+
"grad_norm": 1.4339860677719116,
|
| 1771 |
+
"learning_rate": 4.569083447332422e-05,
|
| 1772 |
+
"loss": 3.1874,
|
| 1773 |
+
"step": 1008
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.09,
|
| 1777 |
+
"grad_norm": 1.5611178874969482,
|
| 1778 |
+
"learning_rate": 4.567373461012312e-05,
|
| 1779 |
+
"loss": 3.075,
|
| 1780 |
+
"step": 1012
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.09,
|
| 1784 |
+
"grad_norm": 1.6089402437210083,
|
| 1785 |
+
"learning_rate": 4.565663474692203e-05,
|
| 1786 |
+
"loss": 3.1544,
|
| 1787 |
+
"step": 1016
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.09,
|
| 1791 |
+
"grad_norm": 1.8874709606170654,
|
| 1792 |
+
"learning_rate": 4.563953488372093e-05,
|
| 1793 |
+
"loss": 3.2806,
|
| 1794 |
+
"step": 1020
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.09,
|
| 1798 |
+
"grad_norm": 1.5624805688858032,
|
| 1799 |
+
"learning_rate": 4.5622435020519835e-05,
|
| 1800 |
+
"loss": 3.1442,
|
| 1801 |
+
"step": 1024
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.09,
|
| 1805 |
+
"grad_norm": 1.968634009361267,
|
| 1806 |
+
"learning_rate": 4.560533515731875e-05,
|
| 1807 |
+
"loss": 3.2743,
|
| 1808 |
+
"step": 1028
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.09,
|
| 1812 |
+
"grad_norm": 1.4064279794692993,
|
| 1813 |
+
"learning_rate": 4.558823529411765e-05,
|
| 1814 |
+
"loss": 3.0726,
|
| 1815 |
+
"step": 1032
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.09,
|
| 1819 |
+
"grad_norm": 1.577048897743225,
|
| 1820 |
+
"learning_rate": 4.557113543091656e-05,
|
| 1821 |
+
"loss": 3.1129,
|
| 1822 |
+
"step": 1036
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.09,
|
| 1826 |
+
"grad_norm": 1.548323392868042,
|
| 1827 |
+
"learning_rate": 4.555403556771546e-05,
|
| 1828 |
+
"loss": 3.144,
|
| 1829 |
+
"step": 1040
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.09,
|
| 1833 |
+
"grad_norm": 1.5542961359024048,
|
| 1834 |
+
"learning_rate": 4.5536935704514365e-05,
|
| 1835 |
+
"loss": 3.182,
|
| 1836 |
+
"step": 1044
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.09,
|
| 1840 |
+
"grad_norm": 1.5506356954574585,
|
| 1841 |
+
"learning_rate": 4.551983584131327e-05,
|
| 1842 |
+
"loss": 3.2186,
|
| 1843 |
+
"step": 1048
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.09,
|
| 1847 |
+
"grad_norm": 1.6962274312973022,
|
| 1848 |
+
"learning_rate": 4.550273597811218e-05,
|
| 1849 |
+
"loss": 3.0385,
|
| 1850 |
+
"step": 1052
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.09,
|
| 1854 |
+
"grad_norm": 1.4382833242416382,
|
| 1855 |
+
"learning_rate": 4.548563611491109e-05,
|
| 1856 |
+
"loss": 3.1268,
|
| 1857 |
+
"step": 1056
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.09,
|
| 1861 |
+
"grad_norm": 2.269392728805542,
|
| 1862 |
+
"learning_rate": 4.546853625170999e-05,
|
| 1863 |
+
"loss": 3.2579,
|
| 1864 |
+
"step": 1060
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.09,
|
| 1868 |
+
"grad_norm": 1.5414201021194458,
|
| 1869 |
+
"learning_rate": 4.5451436388508895e-05,
|
| 1870 |
+
"loss": 3.1501,
|
| 1871 |
+
"step": 1064
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.09,
|
| 1875 |
+
"grad_norm": 1.936946153640747,
|
| 1876 |
+
"learning_rate": 4.54343365253078e-05,
|
| 1877 |
+
"loss": 3.1711,
|
| 1878 |
+
"step": 1068
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.09,
|
| 1882 |
+
"grad_norm": 1.7430529594421387,
|
| 1883 |
+
"learning_rate": 4.54172366621067e-05,
|
| 1884 |
+
"loss": 3.1299,
|
| 1885 |
+
"step": 1072
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.09,
|
| 1889 |
+
"grad_norm": 1.3500404357910156,
|
| 1890 |
+
"learning_rate": 4.540013679890561e-05,
|
| 1891 |
+
"loss": 3.0935,
|
| 1892 |
+
"step": 1076
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.09,
|
| 1896 |
+
"grad_norm": 1.5768132209777832,
|
| 1897 |
+
"learning_rate": 4.538303693570452e-05,
|
| 1898 |
+
"loss": 3.1091,
|
| 1899 |
+
"step": 1080
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.09,
|
| 1903 |
+
"grad_norm": 1.4829493761062622,
|
| 1904 |
+
"learning_rate": 4.5365937072503425e-05,
|
| 1905 |
+
"loss": 3.0862,
|
| 1906 |
+
"step": 1084
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.09,
|
| 1910 |
+
"grad_norm": 1.5560483932495117,
|
| 1911 |
+
"learning_rate": 4.5348837209302326e-05,
|
| 1912 |
+
"loss": 3.1024,
|
| 1913 |
+
"step": 1088
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.09,
|
| 1917 |
+
"grad_norm": 1.6295199394226074,
|
| 1918 |
+
"learning_rate": 4.533173734610123e-05,
|
| 1919 |
+
"loss": 2.9539,
|
| 1920 |
+
"step": 1092
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.09,
|
| 1924 |
+
"grad_norm": 2.1724135875701904,
|
| 1925 |
+
"learning_rate": 4.531463748290014e-05,
|
| 1926 |
+
"loss": 3.1936,
|
| 1927 |
+
"step": 1096
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.09,
|
| 1931 |
+
"grad_norm": 1.7400479316711426,
|
| 1932 |
+
"learning_rate": 4.529753761969904e-05,
|
| 1933 |
+
"loss": 3.1873,
|
| 1934 |
+
"step": 1100
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.09,
|
| 1938 |
+
"grad_norm": 1.554962396621704,
|
| 1939 |
+
"learning_rate": 4.528043775649795e-05,
|
| 1940 |
+
"loss": 3.0643,
|
| 1941 |
+
"step": 1104
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.09,
|
| 1945 |
+
"grad_norm": 1.3722119331359863,
|
| 1946 |
+
"learning_rate": 4.5263337893296855e-05,
|
| 1947 |
+
"loss": 3.0088,
|
| 1948 |
+
"step": 1108
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.1,
|
| 1952 |
+
"grad_norm": 1.395577311515808,
|
| 1953 |
+
"learning_rate": 4.524623803009576e-05,
|
| 1954 |
+
"loss": 3.0584,
|
| 1955 |
+
"step": 1112
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.1,
|
| 1959 |
+
"grad_norm": 1.509710431098938,
|
| 1960 |
+
"learning_rate": 4.522913816689467e-05,
|
| 1961 |
+
"loss": 3.2276,
|
| 1962 |
+
"step": 1116
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.1,
|
| 1966 |
+
"grad_norm": 1.5213786363601685,
|
| 1967 |
+
"learning_rate": 4.521203830369357e-05,
|
| 1968 |
+
"loss": 2.9433,
|
| 1969 |
+
"step": 1120
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.1,
|
| 1973 |
+
"grad_norm": 1.51616632938385,
|
| 1974 |
+
"learning_rate": 4.519493844049248e-05,
|
| 1975 |
+
"loss": 2.9902,
|
| 1976 |
+
"step": 1124
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.1,
|
| 1980 |
+
"grad_norm": 1.5429805517196655,
|
| 1981 |
+
"learning_rate": 4.517783857729138e-05,
|
| 1982 |
+
"loss": 3.0884,
|
| 1983 |
+
"step": 1128
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.1,
|
| 1987 |
+
"grad_norm": 1.4714343547821045,
|
| 1988 |
+
"learning_rate": 4.5160738714090286e-05,
|
| 1989 |
+
"loss": 3.0322,
|
| 1990 |
+
"step": 1132
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.1,
|
| 1994 |
+
"grad_norm": 1.4637647867202759,
|
| 1995 |
+
"learning_rate": 4.51436388508892e-05,
|
| 1996 |
+
"loss": 3.0352,
|
| 1997 |
+
"step": 1136
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.1,
|
| 2001 |
+
"grad_norm": 1.3424347639083862,
|
| 2002 |
+
"learning_rate": 4.51265389876881e-05,
|
| 2003 |
+
"loss": 3.0866,
|
| 2004 |
+
"step": 1140
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.1,
|
| 2008 |
+
"grad_norm": 1.689968228340149,
|
| 2009 |
+
"learning_rate": 4.510943912448701e-05,
|
| 2010 |
+
"loss": 3.0002,
|
| 2011 |
+
"step": 1144
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.1,
|
| 2015 |
+
"grad_norm": 1.9935293197631836,
|
| 2016 |
+
"learning_rate": 4.509233926128591e-05,
|
| 2017 |
+
"loss": 2.9518,
|
| 2018 |
+
"step": 1148
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.1,
|
| 2022 |
+
"grad_norm": 1.5461673736572266,
|
| 2023 |
+
"learning_rate": 4.5075239398084816e-05,
|
| 2024 |
+
"loss": 3.0394,
|
| 2025 |
+
"step": 1152
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.1,
|
| 2029 |
+
"grad_norm": 1.5787737369537354,
|
| 2030 |
+
"learning_rate": 4.505813953488372e-05,
|
| 2031 |
+
"loss": 3.0934,
|
| 2032 |
+
"step": 1156
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.1,
|
| 2036 |
+
"grad_norm": 1.4527121782302856,
|
| 2037 |
+
"learning_rate": 4.504103967168263e-05,
|
| 2038 |
+
"loss": 2.9734,
|
| 2039 |
+
"step": 1160
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.1,
|
| 2043 |
+
"grad_norm": 1.6236604452133179,
|
| 2044 |
+
"learning_rate": 4.502393980848154e-05,
|
| 2045 |
+
"loss": 3.021,
|
| 2046 |
+
"step": 1164
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.1,
|
| 2050 |
+
"grad_norm": 1.5616329908370972,
|
| 2051 |
+
"learning_rate": 4.500683994528044e-05,
|
| 2052 |
+
"loss": 3.0602,
|
| 2053 |
+
"step": 1168
|
| 2054 |
+
}
|
| 2055 |
+
],
|
| 2056 |
+
"logging_steps": 4,
|
| 2057 |
+
"max_steps": 11696,
|
| 2058 |
+
"num_input_tokens_seen": 0,
|
| 2059 |
+
"num_train_epochs": 1,
|
| 2060 |
+
"save_steps": 1170,
|
| 2061 |
+
"total_flos": 1.585392327327744e+16,
|
| 2062 |
+
"train_batch_size": 4,
|
| 2063 |
+
"trial_name": null,
|
| 2064 |
+
"trial_params": null
|
| 2065 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8506ac91feec3f2aac8256c73aae17dbd7cfdd47d79fd5896f5e54f4545690f
|
| 3 |
+
size 4920
|