Upload folder using huggingface_hub
Browse files- checkpoint-11900/added_tokens.json +0 -0
- checkpoint-11900/config.json +45 -0
- checkpoint-11900/model.safetensors +3 -0
- checkpoint-11900/optimizer.pt +3 -0
- checkpoint-11900/rng_state.pth +3 -0
- checkpoint-11900/scaler.pt +3 -0
- checkpoint-11900/scheduler.pt +3 -0
- checkpoint-11900/special_tokens_map.json +7 -0
- checkpoint-11900/tokenizer.json +0 -0
- checkpoint-11900/tokenizer_config.json +0 -0
- checkpoint-11900/trainer_state.json +1718 -0
- checkpoint-11900/training_args.bin +3 -0
- checkpoint-11900/vocab.txt +0 -0
- checkpoint-5950/added_tokens.json +0 -0
- checkpoint-5950/config.json +45 -0
- checkpoint-5950/model.safetensors +3 -0
- checkpoint-5950/optimizer.pt +3 -0
- checkpoint-5950/rng_state.pth +3 -0
- checkpoint-5950/scaler.pt +3 -0
- checkpoint-5950/scheduler.pt +3 -0
- checkpoint-5950/special_tokens_map.json +7 -0
- checkpoint-5950/tokenizer.json +0 -0
- checkpoint-5950/tokenizer_config.json +0 -0
- checkpoint-5950/trainer_state.json +876 -0
- checkpoint-5950/training_args.bin +3 -0
- checkpoint-5950/vocab.txt +0 -0
checkpoint-11900/added_tokens.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-11900/config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"directionality": "bidi",
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0",
|
| 14 |
+
"1": "LABEL_1",
|
| 15 |
+
"2": "LABEL_2",
|
| 16 |
+
"3": "LABEL_3",
|
| 17 |
+
"4": "LABEL_4"
|
| 18 |
+
},
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 3072,
|
| 21 |
+
"label2id": {
|
| 22 |
+
"LABEL_0": 0,
|
| 23 |
+
"LABEL_1": 1,
|
| 24 |
+
"LABEL_2": 2,
|
| 25 |
+
"LABEL_3": 3,
|
| 26 |
+
"LABEL_4": 4
|
| 27 |
+
},
|
| 28 |
+
"layer_norm_eps": 1e-12,
|
| 29 |
+
"max_position_embeddings": 512,
|
| 30 |
+
"model_type": "bert",
|
| 31 |
+
"num_attention_heads": 12,
|
| 32 |
+
"num_hidden_layers": 12,
|
| 33 |
+
"pad_token_id": 0,
|
| 34 |
+
"pooler_fc_size": 768,
|
| 35 |
+
"pooler_num_attention_heads": 12,
|
| 36 |
+
"pooler_num_fc_layers": 3,
|
| 37 |
+
"pooler_size_per_head": 128,
|
| 38 |
+
"pooler_type": "first_token_transform",
|
| 39 |
+
"position_embedding_type": "absolute",
|
| 40 |
+
"problem_type": "single_label_classification",
|
| 41 |
+
"transformers_version": "4.57.3",
|
| 42 |
+
"type_vocab_size": 2,
|
| 43 |
+
"use_cache": true,
|
| 44 |
+
"vocab_size": 135776
|
| 45 |
+
}
|
checkpoint-11900/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7c0811cca83e150a93b08399bb644cb6b31860cccb6f0d0c6eaac267512ed75
|
| 3 |
+
size 761308172
|
checkpoint-11900/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5bbb6efa953ac27ea79f078b7fbfd1ac9747a689ee953824258fca8839b06d1
|
| 3 |
+
size 1522740619
|
checkpoint-11900/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6d3e73ef2de087bb9e086610b317b0edecc499fafca92e7f781b399db6465c5
|
| 3 |
+
size 14645
|
checkpoint-11900/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fca1b27ac585e87f33a001c7dfb0adbdfb56b4d914009fde4324a9b10df26171
|
| 3 |
+
size 1383
|
checkpoint-11900/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:438bdff52caf9cc068adfe9e872400b531669351efd9b7b15a0e9b865935d611
|
| 3 |
+
size 1465
|
checkpoint-11900/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
checkpoint-11900/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-11900/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-11900/trainer_state.json
ADDED
|
@@ -0,0 +1,1718 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 11900,
|
| 3 |
+
"best_metric": 0.7922125924560002,
|
| 4 |
+
"best_model_checkpoint": "pseudo_model/checkpoint-11900",
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 11900,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008403361344537815,
|
| 14 |
+
"grad_norm": 70.3099365234375,
|
| 15 |
+
"learning_rate": 1.991764705882353e-05,
|
| 16 |
+
"loss": 0.0639,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01680672268907563,
|
| 21 |
+
"grad_norm": 0.251952201128006,
|
| 22 |
+
"learning_rate": 1.983361344537815e-05,
|
| 23 |
+
"loss": 0.1373,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.025210084033613446,
|
| 28 |
+
"grad_norm": 54.500118255615234,
|
| 29 |
+
"learning_rate": 1.9749579831932776e-05,
|
| 30 |
+
"loss": 0.187,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03361344537815126,
|
| 35 |
+
"grad_norm": 17.24431037902832,
|
| 36 |
+
"learning_rate": 1.9665546218487398e-05,
|
| 37 |
+
"loss": 0.1268,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04201680672268908,
|
| 42 |
+
"grad_norm": 0.7904719114303589,
|
| 43 |
+
"learning_rate": 1.958151260504202e-05,
|
| 44 |
+
"loss": 0.1147,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05042016806722689,
|
| 49 |
+
"grad_norm": 15.803120613098145,
|
| 50 |
+
"learning_rate": 1.949747899159664e-05,
|
| 51 |
+
"loss": 0.0939,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.058823529411764705,
|
| 56 |
+
"grad_norm": 21.106233596801758,
|
| 57 |
+
"learning_rate": 1.9413445378151262e-05,
|
| 58 |
+
"loss": 0.1392,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06722689075630252,
|
| 63 |
+
"grad_norm": 4.552457809448242,
|
| 64 |
+
"learning_rate": 1.9329411764705883e-05,
|
| 65 |
+
"loss": 0.0754,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07563025210084033,
|
| 70 |
+
"grad_norm": 59.192100524902344,
|
| 71 |
+
"learning_rate": 1.9245378151260505e-05,
|
| 72 |
+
"loss": 0.0864,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08403361344537816,
|
| 77 |
+
"grad_norm": 24.423683166503906,
|
| 78 |
+
"learning_rate": 1.9161344537815126e-05,
|
| 79 |
+
"loss": 0.1462,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09243697478991597,
|
| 84 |
+
"grad_norm": 20.817564010620117,
|
| 85 |
+
"learning_rate": 1.907731092436975e-05,
|
| 86 |
+
"loss": 0.1025,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10084033613445378,
|
| 91 |
+
"grad_norm": 16.254493713378906,
|
| 92 |
+
"learning_rate": 1.8993277310924372e-05,
|
| 93 |
+
"loss": 0.1588,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1092436974789916,
|
| 98 |
+
"grad_norm": 43.414573669433594,
|
| 99 |
+
"learning_rate": 1.8909243697478993e-05,
|
| 100 |
+
"loss": 0.0737,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.11764705882352941,
|
| 105 |
+
"grad_norm": 25.70012664794922,
|
| 106 |
+
"learning_rate": 1.8825210084033615e-05,
|
| 107 |
+
"loss": 0.0638,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.12605042016806722,
|
| 112 |
+
"grad_norm": 4.720924377441406,
|
| 113 |
+
"learning_rate": 1.8741176470588236e-05,
|
| 114 |
+
"loss": 0.1043,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.13445378151260504,
|
| 119 |
+
"grad_norm": 57.87378692626953,
|
| 120 |
+
"learning_rate": 1.8657142857142858e-05,
|
| 121 |
+
"loss": 0.1188,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.14285714285714285,
|
| 126 |
+
"grad_norm": 23.307783126831055,
|
| 127 |
+
"learning_rate": 1.857310924369748e-05,
|
| 128 |
+
"loss": 0.1161,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.15126050420168066,
|
| 133 |
+
"grad_norm": 44.87445831298828,
|
| 134 |
+
"learning_rate": 1.8489075630252104e-05,
|
| 135 |
+
"loss": 0.0917,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.15966386554621848,
|
| 140 |
+
"grad_norm": 14.766839981079102,
|
| 141 |
+
"learning_rate": 1.8405042016806725e-05,
|
| 142 |
+
"loss": 0.1296,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.16806722689075632,
|
| 147 |
+
"grad_norm": 8.999948501586914,
|
| 148 |
+
"learning_rate": 1.8321008403361346e-05,
|
| 149 |
+
"loss": 0.0966,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.17647058823529413,
|
| 154 |
+
"grad_norm": 27.70500946044922,
|
| 155 |
+
"learning_rate": 1.8236974789915968e-05,
|
| 156 |
+
"loss": 0.0847,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.18487394957983194,
|
| 161 |
+
"grad_norm": 45.68607711791992,
|
| 162 |
+
"learning_rate": 1.815294117647059e-05,
|
| 163 |
+
"loss": 0.0774,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.19327731092436976,
|
| 168 |
+
"grad_norm": 9.21766185760498,
|
| 169 |
+
"learning_rate": 1.806890756302521e-05,
|
| 170 |
+
"loss": 0.0832,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.20168067226890757,
|
| 175 |
+
"grad_norm": 84.41997528076172,
|
| 176 |
+
"learning_rate": 1.7984873949579832e-05,
|
| 177 |
+
"loss": 0.1061,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.21008403361344538,
|
| 182 |
+
"grad_norm": 11.963436126708984,
|
| 183 |
+
"learning_rate": 1.7900840336134457e-05,
|
| 184 |
+
"loss": 0.1119,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2184873949579832,
|
| 189 |
+
"grad_norm": 30.294477462768555,
|
| 190 |
+
"learning_rate": 1.7816806722689078e-05,
|
| 191 |
+
"loss": 0.089,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.226890756302521,
|
| 196 |
+
"grad_norm": 0.17758768796920776,
|
| 197 |
+
"learning_rate": 1.77327731092437e-05,
|
| 198 |
+
"loss": 0.0671,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.23529411764705882,
|
| 203 |
+
"grad_norm": 0.24741335213184357,
|
| 204 |
+
"learning_rate": 1.764873949579832e-05,
|
| 205 |
+
"loss": 0.0875,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.24369747899159663,
|
| 210 |
+
"grad_norm": 79.4989013671875,
|
| 211 |
+
"learning_rate": 1.7564705882352942e-05,
|
| 212 |
+
"loss": 0.139,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.25210084033613445,
|
| 217 |
+
"grad_norm": 40.87698745727539,
|
| 218 |
+
"learning_rate": 1.7480672268907564e-05,
|
| 219 |
+
"loss": 0.0666,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2605042016806723,
|
| 224 |
+
"grad_norm": 4.265988826751709,
|
| 225 |
+
"learning_rate": 1.7396638655462185e-05,
|
| 226 |
+
"loss": 0.0957,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2689075630252101,
|
| 231 |
+
"grad_norm": 31.657917022705078,
|
| 232 |
+
"learning_rate": 1.7312605042016806e-05,
|
| 233 |
+
"loss": 0.1013,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2773109243697479,
|
| 238 |
+
"grad_norm": 48.78037643432617,
|
| 239 |
+
"learning_rate": 1.722857142857143e-05,
|
| 240 |
+
"loss": 0.0721,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2857142857142857,
|
| 245 |
+
"grad_norm": 0.1120712086558342,
|
| 246 |
+
"learning_rate": 1.7144537815126052e-05,
|
| 247 |
+
"loss": 0.0865,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.29411764705882354,
|
| 252 |
+
"grad_norm": 0.8699374198913574,
|
| 253 |
+
"learning_rate": 1.7060504201680674e-05,
|
| 254 |
+
"loss": 0.0779,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3025210084033613,
|
| 259 |
+
"grad_norm": 64.08531951904297,
|
| 260 |
+
"learning_rate": 1.6976470588235295e-05,
|
| 261 |
+
"loss": 0.0938,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.31092436974789917,
|
| 266 |
+
"grad_norm": 36.72616195678711,
|
| 267 |
+
"learning_rate": 1.6892436974789917e-05,
|
| 268 |
+
"loss": 0.068,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.31932773109243695,
|
| 273 |
+
"grad_norm": 24.38673973083496,
|
| 274 |
+
"learning_rate": 1.6808403361344538e-05,
|
| 275 |
+
"loss": 0.0667,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3277310924369748,
|
| 280 |
+
"grad_norm": 4.634633541107178,
|
| 281 |
+
"learning_rate": 1.672436974789916e-05,
|
| 282 |
+
"loss": 0.1098,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.33613445378151263,
|
| 287 |
+
"grad_norm": 0.005323654506355524,
|
| 288 |
+
"learning_rate": 1.6640336134453784e-05,
|
| 289 |
+
"loss": 0.0578,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3445378151260504,
|
| 294 |
+
"grad_norm": 30.870010375976562,
|
| 295 |
+
"learning_rate": 1.6556302521008405e-05,
|
| 296 |
+
"loss": 0.0702,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.35294117647058826,
|
| 301 |
+
"grad_norm": 0.3597383201122284,
|
| 302 |
+
"learning_rate": 1.6472268907563027e-05,
|
| 303 |
+
"loss": 0.087,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.36134453781512604,
|
| 308 |
+
"grad_norm": 43.77628707885742,
|
| 309 |
+
"learning_rate": 1.6388235294117648e-05,
|
| 310 |
+
"loss": 0.0865,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3697478991596639,
|
| 315 |
+
"grad_norm": 5.36348819732666,
|
| 316 |
+
"learning_rate": 1.630420168067227e-05,
|
| 317 |
+
"loss": 0.1211,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.37815126050420167,
|
| 322 |
+
"grad_norm": 5.006982326507568,
|
| 323 |
+
"learning_rate": 1.622016806722689e-05,
|
| 324 |
+
"loss": 0.0772,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3865546218487395,
|
| 329 |
+
"grad_norm": 0.21585534512996674,
|
| 330 |
+
"learning_rate": 1.6136134453781512e-05,
|
| 331 |
+
"loss": 0.0873,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3949579831932773,
|
| 336 |
+
"grad_norm": 5.9591193199157715,
|
| 337 |
+
"learning_rate": 1.6052100840336137e-05,
|
| 338 |
+
"loss": 0.0678,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.40336134453781514,
|
| 343 |
+
"grad_norm": 5.950344085693359,
|
| 344 |
+
"learning_rate": 1.596806722689076e-05,
|
| 345 |
+
"loss": 0.0583,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.4117647058823529,
|
| 350 |
+
"grad_norm": 62.205570220947266,
|
| 351 |
+
"learning_rate": 1.588403361344538e-05,
|
| 352 |
+
"loss": 0.0799,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.42016806722689076,
|
| 357 |
+
"grad_norm": 11.344050407409668,
|
| 358 |
+
"learning_rate": 1.58e-05,
|
| 359 |
+
"loss": 0.0925,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.42857142857142855,
|
| 364 |
+
"grad_norm": 20.887693405151367,
|
| 365 |
+
"learning_rate": 1.5715966386554623e-05,
|
| 366 |
+
"loss": 0.0515,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.4369747899159664,
|
| 371 |
+
"grad_norm": 4.148672580718994,
|
| 372 |
+
"learning_rate": 1.5631932773109244e-05,
|
| 373 |
+
"loss": 0.0836,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.44537815126050423,
|
| 378 |
+
"grad_norm": 0.019104987382888794,
|
| 379 |
+
"learning_rate": 1.5547899159663865e-05,
|
| 380 |
+
"loss": 0.0663,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.453781512605042,
|
| 385 |
+
"grad_norm": 55.08736038208008,
|
| 386 |
+
"learning_rate": 1.5463865546218487e-05,
|
| 387 |
+
"loss": 0.0573,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.46218487394957986,
|
| 392 |
+
"grad_norm": 24.77686882019043,
|
| 393 |
+
"learning_rate": 1.537983193277311e-05,
|
| 394 |
+
"loss": 0.0995,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.47058823529411764,
|
| 399 |
+
"grad_norm": 0.2827483117580414,
|
| 400 |
+
"learning_rate": 1.5295798319327733e-05,
|
| 401 |
+
"loss": 0.0931,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4789915966386555,
|
| 406 |
+
"grad_norm": 51.127052307128906,
|
| 407 |
+
"learning_rate": 1.5211764705882354e-05,
|
| 408 |
+
"loss": 0.0674,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.48739495798319327,
|
| 413 |
+
"grad_norm": 28.012380599975586,
|
| 414 |
+
"learning_rate": 1.5127731092436977e-05,
|
| 415 |
+
"loss": 0.071,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.4957983193277311,
|
| 420 |
+
"grad_norm": 62.96744155883789,
|
| 421 |
+
"learning_rate": 1.5043697478991597e-05,
|
| 422 |
+
"loss": 0.0802,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5042016806722689,
|
| 427 |
+
"grad_norm": 24.443973541259766,
|
| 428 |
+
"learning_rate": 1.4959663865546218e-05,
|
| 429 |
+
"loss": 0.0822,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5126050420168067,
|
| 434 |
+
"grad_norm": 0.09382598847150803,
|
| 435 |
+
"learning_rate": 1.4875630252100841e-05,
|
| 436 |
+
"loss": 0.0893,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5210084033613446,
|
| 441 |
+
"grad_norm": 8.969447135925293,
|
| 442 |
+
"learning_rate": 1.4791596638655463e-05,
|
| 443 |
+
"loss": 0.0587,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5294117647058824,
|
| 448 |
+
"grad_norm": 0.33123093843460083,
|
| 449 |
+
"learning_rate": 1.4707563025210086e-05,
|
| 450 |
+
"loss": 0.0662,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5378151260504201,
|
| 455 |
+
"grad_norm": 0.34799447655677795,
|
| 456 |
+
"learning_rate": 1.4623529411764707e-05,
|
| 457 |
+
"loss": 0.0471,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5462184873949579,
|
| 462 |
+
"grad_norm": 0.10502827167510986,
|
| 463 |
+
"learning_rate": 1.453949579831933e-05,
|
| 464 |
+
"loss": 0.0608,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5546218487394958,
|
| 469 |
+
"grad_norm": 46.54685974121094,
|
| 470 |
+
"learning_rate": 1.4455462184873952e-05,
|
| 471 |
+
"loss": 0.0756,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5630252100840336,
|
| 476 |
+
"grad_norm": 25.307750701904297,
|
| 477 |
+
"learning_rate": 1.4371428571428571e-05,
|
| 478 |
+
"loss": 0.0703,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5714285714285714,
|
| 483 |
+
"grad_norm": 0.03593587502837181,
|
| 484 |
+
"learning_rate": 1.4287394957983194e-05,
|
| 485 |
+
"loss": 0.0509,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5798319327731093,
|
| 490 |
+
"grad_norm": 6.873576641082764,
|
| 491 |
+
"learning_rate": 1.4203361344537816e-05,
|
| 492 |
+
"loss": 0.0751,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5882352941176471,
|
| 497 |
+
"grad_norm": 39.44701385498047,
|
| 498 |
+
"learning_rate": 1.4119327731092439e-05,
|
| 499 |
+
"loss": 0.0698,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5966386554621849,
|
| 504 |
+
"grad_norm": 0.7871583104133606,
|
| 505 |
+
"learning_rate": 1.403529411764706e-05,
|
| 506 |
+
"loss": 0.0701,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6050420168067226,
|
| 511 |
+
"grad_norm": 0.06720981746912003,
|
| 512 |
+
"learning_rate": 1.3951260504201683e-05,
|
| 513 |
+
"loss": 0.0613,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6134453781512605,
|
| 518 |
+
"grad_norm": 87.64744567871094,
|
| 519 |
+
"learning_rate": 1.3867226890756305e-05,
|
| 520 |
+
"loss": 0.0657,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6218487394957983,
|
| 525 |
+
"grad_norm": 92.26810455322266,
|
| 526 |
+
"learning_rate": 1.3783193277310924e-05,
|
| 527 |
+
"loss": 0.0915,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6302521008403361,
|
| 532 |
+
"grad_norm": 0.13392935693264008,
|
| 533 |
+
"learning_rate": 1.3699159663865547e-05,
|
| 534 |
+
"loss": 0.0552,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6386554621848739,
|
| 539 |
+
"grad_norm": 0.2624678313732147,
|
| 540 |
+
"learning_rate": 1.3615126050420169e-05,
|
| 541 |
+
"loss": 0.061,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6470588235294118,
|
| 546 |
+
"grad_norm": 58.510833740234375,
|
| 547 |
+
"learning_rate": 1.353109243697479e-05,
|
| 548 |
+
"loss": 0.0475,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6554621848739496,
|
| 553 |
+
"grad_norm": 0.19354496896266937,
|
| 554 |
+
"learning_rate": 1.3447058823529413e-05,
|
| 555 |
+
"loss": 0.0423,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6638655462184874,
|
| 560 |
+
"grad_norm": 20.584426879882812,
|
| 561 |
+
"learning_rate": 1.3363025210084035e-05,
|
| 562 |
+
"loss": 0.1135,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6722689075630253,
|
| 567 |
+
"grad_norm": 1.4565203189849854,
|
| 568 |
+
"learning_rate": 1.3278991596638658e-05,
|
| 569 |
+
"loss": 0.0264,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.680672268907563,
|
| 574 |
+
"grad_norm": 0.0815153419971466,
|
| 575 |
+
"learning_rate": 1.3194957983193279e-05,
|
| 576 |
+
"loss": 0.0956,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6890756302521008,
|
| 581 |
+
"grad_norm": 45.101993560791016,
|
| 582 |
+
"learning_rate": 1.3110924369747899e-05,
|
| 583 |
+
"loss": 0.0853,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6974789915966386,
|
| 588 |
+
"grad_norm": 8.502076148986816,
|
| 589 |
+
"learning_rate": 1.3026890756302522e-05,
|
| 590 |
+
"loss": 0.0407,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7058823529411765,
|
| 595 |
+
"grad_norm": 0.01488969475030899,
|
| 596 |
+
"learning_rate": 1.2942857142857143e-05,
|
| 597 |
+
"loss": 0.0506,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7142857142857143,
|
| 602 |
+
"grad_norm": 128.31639099121094,
|
| 603 |
+
"learning_rate": 1.2858823529411766e-05,
|
| 604 |
+
"loss": 0.0527,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.7226890756302521,
|
| 609 |
+
"grad_norm": 69.5967025756836,
|
| 610 |
+
"learning_rate": 1.2774789915966388e-05,
|
| 611 |
+
"loss": 0.0437,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7310924369747899,
|
| 616 |
+
"grad_norm": 51.27083206176758,
|
| 617 |
+
"learning_rate": 1.269075630252101e-05,
|
| 618 |
+
"loss": 0.0511,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7394957983193278,
|
| 623 |
+
"grad_norm": 1.9427380561828613,
|
| 624 |
+
"learning_rate": 1.2606722689075632e-05,
|
| 625 |
+
"loss": 0.0938,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7478991596638656,
|
| 630 |
+
"grad_norm": 17.49405288696289,
|
| 631 |
+
"learning_rate": 1.2522689075630252e-05,
|
| 632 |
+
"loss": 0.0799,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7563025210084033,
|
| 637 |
+
"grad_norm": 2.69028377532959,
|
| 638 |
+
"learning_rate": 1.2438655462184875e-05,
|
| 639 |
+
"loss": 0.0556,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7647058823529411,
|
| 644 |
+
"grad_norm": 91.98591613769531,
|
| 645 |
+
"learning_rate": 1.2354621848739496e-05,
|
| 646 |
+
"loss": 0.0341,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.773109243697479,
|
| 651 |
+
"grad_norm": 37.63540267944336,
|
| 652 |
+
"learning_rate": 1.2270588235294119e-05,
|
| 653 |
+
"loss": 0.0748,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7815126050420168,
|
| 658 |
+
"grad_norm": 24.502737045288086,
|
| 659 |
+
"learning_rate": 1.218655462184874e-05,
|
| 660 |
+
"loss": 0.0693,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7899159663865546,
|
| 665 |
+
"grad_norm": 0.010893125087022781,
|
| 666 |
+
"learning_rate": 1.2102521008403364e-05,
|
| 667 |
+
"loss": 0.0371,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7983193277310925,
|
| 672 |
+
"grad_norm": 39.78943634033203,
|
| 673 |
+
"learning_rate": 1.2018487394957985e-05,
|
| 674 |
+
"loss": 0.0547,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.8067226890756303,
|
| 679 |
+
"grad_norm": 21.680444717407227,
|
| 680 |
+
"learning_rate": 1.1934453781512605e-05,
|
| 681 |
+
"loss": 0.0405,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8151260504201681,
|
| 686 |
+
"grad_norm": 91.88570404052734,
|
| 687 |
+
"learning_rate": 1.1850420168067228e-05,
|
| 688 |
+
"loss": 0.048,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8235294117647058,
|
| 693 |
+
"grad_norm": 47.21152114868164,
|
| 694 |
+
"learning_rate": 1.1766386554621849e-05,
|
| 695 |
+
"loss": 0.0957,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8319327731092437,
|
| 700 |
+
"grad_norm": 0.03324070945382118,
|
| 701 |
+
"learning_rate": 1.1682352941176472e-05,
|
| 702 |
+
"loss": 0.0525,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8403361344537815,
|
| 707 |
+
"grad_norm": 0.4629770517349243,
|
| 708 |
+
"learning_rate": 1.1598319327731094e-05,
|
| 709 |
+
"loss": 0.0682,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8487394957983193,
|
| 714 |
+
"grad_norm": 45.92827224731445,
|
| 715 |
+
"learning_rate": 1.1514285714285715e-05,
|
| 716 |
+
"loss": 0.0424,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8571428571428571,
|
| 721 |
+
"grad_norm": 32.47571563720703,
|
| 722 |
+
"learning_rate": 1.1430252100840338e-05,
|
| 723 |
+
"loss": 0.0794,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.865546218487395,
|
| 728 |
+
"grad_norm": 17.017562866210938,
|
| 729 |
+
"learning_rate": 1.134621848739496e-05,
|
| 730 |
+
"loss": 0.0436,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8739495798319328,
|
| 735 |
+
"grad_norm": 2.4690027236938477,
|
| 736 |
+
"learning_rate": 1.1262184873949579e-05,
|
| 737 |
+
"loss": 0.0618,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8823529411764706,
|
| 742 |
+
"grad_norm": 91.57916259765625,
|
| 743 |
+
"learning_rate": 1.1178151260504202e-05,
|
| 744 |
+
"loss": 0.0628,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8907563025210085,
|
| 749 |
+
"grad_norm": 0.11016395688056946,
|
| 750 |
+
"learning_rate": 1.1094117647058823e-05,
|
| 751 |
+
"loss": 0.076,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8991596638655462,
|
| 756 |
+
"grad_norm": 42.38503646850586,
|
| 757 |
+
"learning_rate": 1.1010084033613447e-05,
|
| 758 |
+
"loss": 0.0759,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.907563025210084,
|
| 763 |
+
"grad_norm": 4.813717365264893,
|
| 764 |
+
"learning_rate": 1.0926050420168068e-05,
|
| 765 |
+
"loss": 0.0349,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.9159663865546218,
|
| 770 |
+
"grad_norm": 0.016100173816084862,
|
| 771 |
+
"learning_rate": 1.0842016806722691e-05,
|
| 772 |
+
"loss": 0.0506,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.9243697478991597,
|
| 777 |
+
"grad_norm": 0.07725575566291809,
|
| 778 |
+
"learning_rate": 1.0757983193277312e-05,
|
| 779 |
+
"loss": 0.0581,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9327731092436975,
|
| 784 |
+
"grad_norm": 69.73644256591797,
|
| 785 |
+
"learning_rate": 1.0673949579831932e-05,
|
| 786 |
+
"loss": 0.0552,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9411764705882353,
|
| 791 |
+
"grad_norm": 0.0879245325922966,
|
| 792 |
+
"learning_rate": 1.0589915966386555e-05,
|
| 793 |
+
"loss": 0.059,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9495798319327731,
|
| 798 |
+
"grad_norm": 0.06376896053552628,
|
| 799 |
+
"learning_rate": 1.0505882352941176e-05,
|
| 800 |
+
"loss": 0.046,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.957983193277311,
|
| 805 |
+
"grad_norm": 98.75856018066406,
|
| 806 |
+
"learning_rate": 1.04218487394958e-05,
|
| 807 |
+
"loss": 0.0319,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9663865546218487,
|
| 812 |
+
"grad_norm": 0.07460715621709824,
|
| 813 |
+
"learning_rate": 1.0337815126050421e-05,
|
| 814 |
+
"loss": 0.0349,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9747899159663865,
|
| 819 |
+
"grad_norm": 88.64122772216797,
|
| 820 |
+
"learning_rate": 1.0253781512605044e-05,
|
| 821 |
+
"loss": 0.0577,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9831932773109243,
|
| 826 |
+
"grad_norm": 0.0315723791718483,
|
| 827 |
+
"learning_rate": 1.0169747899159665e-05,
|
| 828 |
+
"loss": 0.0655,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9915966386554622,
|
| 833 |
+
"grad_norm": 74.6689682006836,
|
| 834 |
+
"learning_rate": 1.0085714285714288e-05,
|
| 835 |
+
"loss": 0.0457,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.0,
|
| 840 |
+
"grad_norm": 0.06222504749894142,
|
| 841 |
+
"learning_rate": 1.0001680672268908e-05,
|
| 842 |
+
"loss": 0.0391,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.0,
|
| 847 |
+
"eval_f1": 0.7729265244559362,
|
| 848 |
+
"eval_loss": 1.5595977306365967,
|
| 849 |
+
"eval_runtime": 0.0922,
|
| 850 |
+
"eval_samples_per_second": 2364.65,
|
| 851 |
+
"eval_steps_per_second": 43.388,
|
| 852 |
+
"step": 5950
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 1.0084033613445378,
|
| 856 |
+
"grad_norm": 0.01488153450191021,
|
| 857 |
+
"learning_rate": 9.917647058823531e-06,
|
| 858 |
+
"loss": 0.0245,
|
| 859 |
+
"step": 6000
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 1.0168067226890756,
|
| 863 |
+
"grad_norm": 0.013586599379777908,
|
| 864 |
+
"learning_rate": 9.833613445378152e-06,
|
| 865 |
+
"loss": 0.0063,
|
| 866 |
+
"step": 6050
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.0252100840336134,
|
| 870 |
+
"grad_norm": 0.6163426041603088,
|
| 871 |
+
"learning_rate": 9.749579831932774e-06,
|
| 872 |
+
"loss": 0.0394,
|
| 873 |
+
"step": 6100
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 1.0336134453781514,
|
| 877 |
+
"grad_norm": 14.39757251739502,
|
| 878 |
+
"learning_rate": 9.665546218487395e-06,
|
| 879 |
+
"loss": 0.013,
|
| 880 |
+
"step": 6150
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 1.0420168067226891,
|
| 884 |
+
"grad_norm": 0.016938969492912292,
|
| 885 |
+
"learning_rate": 9.581512605042017e-06,
|
| 886 |
+
"loss": 0.0265,
|
| 887 |
+
"step": 6200
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 1.050420168067227,
|
| 891 |
+
"grad_norm": 0.5566484928131104,
|
| 892 |
+
"learning_rate": 9.49747899159664e-06,
|
| 893 |
+
"loss": 0.0145,
|
| 894 |
+
"step": 6250
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.0588235294117647,
|
| 898 |
+
"grad_norm": 0.004737951327115297,
|
| 899 |
+
"learning_rate": 9.413445378151261e-06,
|
| 900 |
+
"loss": 0.0306,
|
| 901 |
+
"step": 6300
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 1.0672268907563025,
|
| 905 |
+
"grad_norm": 0.6328762173652649,
|
| 906 |
+
"learning_rate": 9.329411764705884e-06,
|
| 907 |
+
"loss": 0.0247,
|
| 908 |
+
"step": 6350
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 1.0756302521008403,
|
| 912 |
+
"grad_norm": 0.003608213271945715,
|
| 913 |
+
"learning_rate": 9.245378151260504e-06,
|
| 914 |
+
"loss": 0.0293,
|
| 915 |
+
"step": 6400
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 1.084033613445378,
|
| 919 |
+
"grad_norm": 0.7948128581047058,
|
| 920 |
+
"learning_rate": 9.161344537815127e-06,
|
| 921 |
+
"loss": 0.0374,
|
| 922 |
+
"step": 6450
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.092436974789916,
|
| 926 |
+
"grad_norm": 0.048644352704286575,
|
| 927 |
+
"learning_rate": 9.077310924369748e-06,
|
| 928 |
+
"loss": 0.0237,
|
| 929 |
+
"step": 6500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.1008403361344539,
|
| 933 |
+
"grad_norm": 0.0088767996057868,
|
| 934 |
+
"learning_rate": 8.993277310924371e-06,
|
| 935 |
+
"loss": 0.021,
|
| 936 |
+
"step": 6550
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.1092436974789917,
|
| 940 |
+
"grad_norm": 0.022422347217798233,
|
| 941 |
+
"learning_rate": 8.909243697478993e-06,
|
| 942 |
+
"loss": 0.025,
|
| 943 |
+
"step": 6600
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 1.1176470588235294,
|
| 947 |
+
"grad_norm": 1.32077956199646,
|
| 948 |
+
"learning_rate": 8.825210084033614e-06,
|
| 949 |
+
"loss": 0.0176,
|
| 950 |
+
"step": 6650
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 1.1260504201680672,
|
| 954 |
+
"grad_norm": 0.0784287080168724,
|
| 955 |
+
"learning_rate": 8.741176470588237e-06,
|
| 956 |
+
"loss": 0.0299,
|
| 957 |
+
"step": 6700
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 1.134453781512605,
|
| 961 |
+
"grad_norm": 96.36627197265625,
|
| 962 |
+
"learning_rate": 8.657142857142858e-06,
|
| 963 |
+
"loss": 0.023,
|
| 964 |
+
"step": 6750
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 1.1428571428571428,
|
| 968 |
+
"grad_norm": 16.85433006286621,
|
| 969 |
+
"learning_rate": 8.57310924369748e-06,
|
| 970 |
+
"loss": 0.0448,
|
| 971 |
+
"step": 6800
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 1.1512605042016806,
|
| 975 |
+
"grad_norm": 0.02477632649242878,
|
| 976 |
+
"learning_rate": 8.489075630252101e-06,
|
| 977 |
+
"loss": 0.0108,
|
| 978 |
+
"step": 6850
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 1.1596638655462184,
|
| 982 |
+
"grad_norm": 0.0038683288730680943,
|
| 983 |
+
"learning_rate": 8.405042016806724e-06,
|
| 984 |
+
"loss": 0.0232,
|
| 985 |
+
"step": 6900
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 1.1680672268907564,
|
| 989 |
+
"grad_norm": 14.223528861999512,
|
| 990 |
+
"learning_rate": 8.321008403361344e-06,
|
| 991 |
+
"loss": 0.031,
|
| 992 |
+
"step": 6950
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 1.1764705882352942,
|
| 996 |
+
"grad_norm": 0.008225447498261929,
|
| 997 |
+
"learning_rate": 8.236974789915967e-06,
|
| 998 |
+
"loss": 0.0315,
|
| 999 |
+
"step": 7000
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 1.184873949579832,
|
| 1003 |
+
"grad_norm": 0.1930830180644989,
|
| 1004 |
+
"learning_rate": 8.152941176470588e-06,
|
| 1005 |
+
"loss": 0.0205,
|
| 1006 |
+
"step": 7050
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.1932773109243697,
|
| 1010 |
+
"grad_norm": 0.003636949462816119,
|
| 1011 |
+
"learning_rate": 8.068907563025211e-06,
|
| 1012 |
+
"loss": 0.0093,
|
| 1013 |
+
"step": 7100
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 1.2016806722689075,
|
| 1017 |
+
"grad_norm": 0.0013511282159015536,
|
| 1018 |
+
"learning_rate": 7.984873949579833e-06,
|
| 1019 |
+
"loss": 0.0251,
|
| 1020 |
+
"step": 7150
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 1.2100840336134453,
|
| 1024 |
+
"grad_norm": 0.011679396033287048,
|
| 1025 |
+
"learning_rate": 7.900840336134454e-06,
|
| 1026 |
+
"loss": 0.0193,
|
| 1027 |
+
"step": 7200
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 1.2184873949579833,
|
| 1031 |
+
"grad_norm": 0.01257317140698433,
|
| 1032 |
+
"learning_rate": 7.816806722689077e-06,
|
| 1033 |
+
"loss": 0.0088,
|
| 1034 |
+
"step": 7250
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.226890756302521,
|
| 1038 |
+
"grad_norm": 0.004438911098986864,
|
| 1039 |
+
"learning_rate": 7.732773109243699e-06,
|
| 1040 |
+
"loss": 0.0196,
|
| 1041 |
+
"step": 7300
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 1.2352941176470589,
|
| 1045 |
+
"grad_norm": 0.005053033120930195,
|
| 1046 |
+
"learning_rate": 7.64873949579832e-06,
|
| 1047 |
+
"loss": 0.0337,
|
| 1048 |
+
"step": 7350
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.2436974789915967,
|
| 1052 |
+
"grad_norm": 0.013792096637189388,
|
| 1053 |
+
"learning_rate": 7.564705882352941e-06,
|
| 1054 |
+
"loss": 0.0085,
|
| 1055 |
+
"step": 7400
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 1.2521008403361344,
|
| 1059 |
+
"grad_norm": 0.004068428184837103,
|
| 1060 |
+
"learning_rate": 7.480672268907564e-06,
|
| 1061 |
+
"loss": 0.0292,
|
| 1062 |
+
"step": 7450
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 1.2605042016806722,
|
| 1066 |
+
"grad_norm": 0.09173774719238281,
|
| 1067 |
+
"learning_rate": 7.396638655462185e-06,
|
| 1068 |
+
"loss": 0.0193,
|
| 1069 |
+
"step": 7500
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 1.26890756302521,
|
| 1073 |
+
"grad_norm": 70.79894256591797,
|
| 1074 |
+
"learning_rate": 7.312605042016807e-06,
|
| 1075 |
+
"loss": 0.0232,
|
| 1076 |
+
"step": 7550
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.2773109243697478,
|
| 1080 |
+
"grad_norm": 0.8466151356697083,
|
| 1081 |
+
"learning_rate": 7.2285714285714294e-06,
|
| 1082 |
+
"loss": 0.0053,
|
| 1083 |
+
"step": 7600
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 1.2857142857142856,
|
| 1087 |
+
"grad_norm": 0.008977376855909824,
|
| 1088 |
+
"learning_rate": 7.144537815126052e-06,
|
| 1089 |
+
"loss": 0.0349,
|
| 1090 |
+
"step": 7650
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.2941176470588236,
|
| 1094 |
+
"grad_norm": 43.390647888183594,
|
| 1095 |
+
"learning_rate": 7.060504201680672e-06,
|
| 1096 |
+
"loss": 0.0264,
|
| 1097 |
+
"step": 7700
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 1.3025210084033614,
|
| 1101 |
+
"grad_norm": 24.95809555053711,
|
| 1102 |
+
"learning_rate": 6.976470588235294e-06,
|
| 1103 |
+
"loss": 0.0213,
|
| 1104 |
+
"step": 7750
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 1.3109243697478992,
|
| 1108 |
+
"grad_norm": 0.033690981566905975,
|
| 1109 |
+
"learning_rate": 6.892436974789917e-06,
|
| 1110 |
+
"loss": 0.021,
|
| 1111 |
+
"step": 7800
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 1.319327731092437,
|
| 1115 |
+
"grad_norm": 0.0035024620592594147,
|
| 1116 |
+
"learning_rate": 6.808403361344539e-06,
|
| 1117 |
+
"loss": 0.0236,
|
| 1118 |
+
"step": 7850
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.3277310924369747,
|
| 1122 |
+
"grad_norm": 0.007407465483993292,
|
| 1123 |
+
"learning_rate": 6.72436974789916e-06,
|
| 1124 |
+
"loss": 0.0238,
|
| 1125 |
+
"step": 7900
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 1.3361344537815127,
|
| 1129 |
+
"grad_norm": 60.801025390625,
|
| 1130 |
+
"learning_rate": 6.640336134453782e-06,
|
| 1131 |
+
"loss": 0.0203,
|
| 1132 |
+
"step": 7950
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 1.3445378151260505,
|
| 1136 |
+
"grad_norm": 0.013495533727109432,
|
| 1137 |
+
"learning_rate": 6.556302521008404e-06,
|
| 1138 |
+
"loss": 0.0319,
|
| 1139 |
+
"step": 8000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 1.3529411764705883,
|
| 1143 |
+
"grad_norm": 0.25706109404563904,
|
| 1144 |
+
"learning_rate": 6.472268907563026e-06,
|
| 1145 |
+
"loss": 0.0313,
|
| 1146 |
+
"step": 8050
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 1.361344537815126,
|
| 1150 |
+
"grad_norm": 37.97319412231445,
|
| 1151 |
+
"learning_rate": 6.388235294117647e-06,
|
| 1152 |
+
"loss": 0.02,
|
| 1153 |
+
"step": 8100
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.3697478991596639,
|
| 1157 |
+
"grad_norm": 0.0020468737930059433,
|
| 1158 |
+
"learning_rate": 6.30420168067227e-06,
|
| 1159 |
+
"loss": 0.0192,
|
| 1160 |
+
"step": 8150
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.3781512605042017,
|
| 1164 |
+
"grad_norm": 0.000602549931500107,
|
| 1165 |
+
"learning_rate": 6.220168067226892e-06,
|
| 1166 |
+
"loss": 0.0201,
|
| 1167 |
+
"step": 8200
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 1.3865546218487395,
|
| 1171 |
+
"grad_norm": 0.07371534407138824,
|
| 1172 |
+
"learning_rate": 6.136134453781512e-06,
|
| 1173 |
+
"loss": 0.0176,
|
| 1174 |
+
"step": 8250
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 1.3949579831932772,
|
| 1178 |
+
"grad_norm": 0.0059112804010510445,
|
| 1179 |
+
"learning_rate": 6.0521008403361346e-06,
|
| 1180 |
+
"loss": 0.0307,
|
| 1181 |
+
"step": 8300
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 1.403361344537815,
|
| 1185 |
+
"grad_norm": 0.00568784074857831,
|
| 1186 |
+
"learning_rate": 5.968067226890757e-06,
|
| 1187 |
+
"loss": 0.0302,
|
| 1188 |
+
"step": 8350
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 1.4117647058823528,
|
| 1192 |
+
"grad_norm": 0.0017624880420044065,
|
| 1193 |
+
"learning_rate": 5.884033613445379e-06,
|
| 1194 |
+
"loss": 0.0265,
|
| 1195 |
+
"step": 8400
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 1.4201680672268908,
|
| 1199 |
+
"grad_norm": 64.0350341796875,
|
| 1200 |
+
"learning_rate": 5.8e-06,
|
| 1201 |
+
"loss": 0.0192,
|
| 1202 |
+
"step": 8450
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 1.4285714285714286,
|
| 1206 |
+
"grad_norm": 0.0009836215758696198,
|
| 1207 |
+
"learning_rate": 5.715966386554623e-06,
|
| 1208 |
+
"loss": 0.0209,
|
| 1209 |
+
"step": 8500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 1.4369747899159664,
|
| 1213 |
+
"grad_norm": 0.0020469457376748323,
|
| 1214 |
+
"learning_rate": 5.631932773109244e-06,
|
| 1215 |
+
"loss": 0.0219,
|
| 1216 |
+
"step": 8550
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 1.4453781512605042,
|
| 1220 |
+
"grad_norm": 0.00875889416784048,
|
| 1221 |
+
"learning_rate": 5.547899159663866e-06,
|
| 1222 |
+
"loss": 0.0248,
|
| 1223 |
+
"step": 8600
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 1.453781512605042,
|
| 1227 |
+
"grad_norm": 5.378687381744385,
|
| 1228 |
+
"learning_rate": 5.4638655462184876e-06,
|
| 1229 |
+
"loss": 0.033,
|
| 1230 |
+
"step": 8650
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 1.46218487394958,
|
| 1234 |
+
"grad_norm": 0.007630588486790657,
|
| 1235 |
+
"learning_rate": 5.37983193277311e-06,
|
| 1236 |
+
"loss": 0.0116,
|
| 1237 |
+
"step": 8700
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 1.4705882352941178,
|
| 1241 |
+
"grad_norm": 0.04181138053536415,
|
| 1242 |
+
"learning_rate": 5.295798319327732e-06,
|
| 1243 |
+
"loss": 0.0216,
|
| 1244 |
+
"step": 8750
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 1.4789915966386555,
|
| 1248 |
+
"grad_norm": 0.005822291597723961,
|
| 1249 |
+
"learning_rate": 5.2117647058823525e-06,
|
| 1250 |
+
"loss": 0.0202,
|
| 1251 |
+
"step": 8800
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 1.4873949579831933,
|
| 1255 |
+
"grad_norm": 0.0021654930897057056,
|
| 1256 |
+
"learning_rate": 5.127731092436975e-06,
|
| 1257 |
+
"loss": 0.0109,
|
| 1258 |
+
"step": 8850
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 1.495798319327731,
|
| 1262 |
+
"grad_norm": 0.22193148732185364,
|
| 1263 |
+
"learning_rate": 5.043697478991597e-06,
|
| 1264 |
+
"loss": 0.0085,
|
| 1265 |
+
"step": 8900
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 1.504201680672269,
|
| 1269 |
+
"grad_norm": 0.0038606910966336727,
|
| 1270 |
+
"learning_rate": 4.959663865546218e-06,
|
| 1271 |
+
"loss": 0.0091,
|
| 1272 |
+
"step": 8950
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 1.5126050420168067,
|
| 1276 |
+
"grad_norm": 0.0015581476036459208,
|
| 1277 |
+
"learning_rate": 4.8756302521008405e-06,
|
| 1278 |
+
"loss": 0.0239,
|
| 1279 |
+
"step": 9000
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 1.5210084033613445,
|
| 1283 |
+
"grad_norm": 0.0007730493089184165,
|
| 1284 |
+
"learning_rate": 4.791596638655463e-06,
|
| 1285 |
+
"loss": 0.0058,
|
| 1286 |
+
"step": 9050
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 1.5294117647058822,
|
| 1290 |
+
"grad_norm": 0.017255844548344612,
|
| 1291 |
+
"learning_rate": 4.707563025210085e-06,
|
| 1292 |
+
"loss": 0.0197,
|
| 1293 |
+
"step": 9100
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 1.53781512605042,
|
| 1297 |
+
"grad_norm": 0.12647680938243866,
|
| 1298 |
+
"learning_rate": 4.623529411764706e-06,
|
| 1299 |
+
"loss": 0.0372,
|
| 1300 |
+
"step": 9150
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 1.5462184873949578,
|
| 1304 |
+
"grad_norm": 73.29251861572266,
|
| 1305 |
+
"learning_rate": 4.539495798319329e-06,
|
| 1306 |
+
"loss": 0.0299,
|
| 1307 |
+
"step": 9200
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.5546218487394958,
|
| 1311 |
+
"grad_norm": 0.2538398206233978,
|
| 1312 |
+
"learning_rate": 4.45546218487395e-06,
|
| 1313 |
+
"loss": 0.0103,
|
| 1314 |
+
"step": 9250
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 1.5630252100840336,
|
| 1318 |
+
"grad_norm": 0.31796854734420776,
|
| 1319 |
+
"learning_rate": 4.371428571428572e-06,
|
| 1320 |
+
"loss": 0.0159,
|
| 1321 |
+
"step": 9300
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 1.5714285714285714,
|
| 1325 |
+
"grad_norm": 0.010294510051608086,
|
| 1326 |
+
"learning_rate": 4.2873949579831935e-06,
|
| 1327 |
+
"loss": 0.0272,
|
| 1328 |
+
"step": 9350
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 1.5798319327731094,
|
| 1332 |
+
"grad_norm": 0.07475843280553818,
|
| 1333 |
+
"learning_rate": 4.203361344537816e-06,
|
| 1334 |
+
"loss": 0.0203,
|
| 1335 |
+
"step": 9400
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 1.5882352941176472,
|
| 1339 |
+
"grad_norm": 1.730021357536316,
|
| 1340 |
+
"learning_rate": 4.119327731092437e-06,
|
| 1341 |
+
"loss": 0.0114,
|
| 1342 |
+
"step": 9450
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 1.596638655462185,
|
| 1346 |
+
"grad_norm": 0.2311408966779709,
|
| 1347 |
+
"learning_rate": 4.0352941176470585e-06,
|
| 1348 |
+
"loss": 0.0124,
|
| 1349 |
+
"step": 9500
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.6050420168067228,
|
| 1353 |
+
"grad_norm": 0.01608281396329403,
|
| 1354 |
+
"learning_rate": 3.951260504201681e-06,
|
| 1355 |
+
"loss": 0.031,
|
| 1356 |
+
"step": 9550
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 1.6134453781512605,
|
| 1360 |
+
"grad_norm": 0.0010955685283988714,
|
| 1361 |
+
"learning_rate": 3.867226890756303e-06,
|
| 1362 |
+
"loss": 0.0339,
|
| 1363 |
+
"step": 9600
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 1.6218487394957983,
|
| 1367 |
+
"grad_norm": 0.0016479927580803633,
|
| 1368 |
+
"learning_rate": 3.7831932773109247e-06,
|
| 1369 |
+
"loss": 0.014,
|
| 1370 |
+
"step": 9650
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 1.6302521008403361,
|
| 1374 |
+
"grad_norm": 0.0014391638105735183,
|
| 1375 |
+
"learning_rate": 3.6991596638655465e-06,
|
| 1376 |
+
"loss": 0.0145,
|
| 1377 |
+
"step": 9700
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.638655462184874,
|
| 1381 |
+
"grad_norm": 0.0035242035519331694,
|
| 1382 |
+
"learning_rate": 3.6151260504201683e-06,
|
| 1383 |
+
"loss": 0.0179,
|
| 1384 |
+
"step": 9750
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.6470588235294117,
|
| 1388 |
+
"grad_norm": 0.0006413786322809756,
|
| 1389 |
+
"learning_rate": 3.53109243697479e-06,
|
| 1390 |
+
"loss": 0.0299,
|
| 1391 |
+
"step": 9800
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 1.6554621848739495,
|
| 1395 |
+
"grad_norm": 0.00946138147264719,
|
| 1396 |
+
"learning_rate": 3.4470588235294123e-06,
|
| 1397 |
+
"loss": 0.0192,
|
| 1398 |
+
"step": 9850
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 1.6638655462184873,
|
| 1402 |
+
"grad_norm": Infinity,
|
| 1403 |
+
"learning_rate": 3.3630252100840337e-06,
|
| 1404 |
+
"loss": 0.0125,
|
| 1405 |
+
"step": 9900
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 1.6722689075630253,
|
| 1409 |
+
"grad_norm": 0.0018359735840931535,
|
| 1410 |
+
"learning_rate": 3.278991596638656e-06,
|
| 1411 |
+
"loss": 0.0068,
|
| 1412 |
+
"step": 9950
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 1.680672268907563,
|
| 1416 |
+
"grad_norm": 9.746746063232422,
|
| 1417 |
+
"learning_rate": 3.1949579831932777e-06,
|
| 1418 |
+
"loss": 0.018,
|
| 1419 |
+
"step": 10000
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 1.6890756302521008,
|
| 1423 |
+
"grad_norm": 0.0029287682846188545,
|
| 1424 |
+
"learning_rate": 3.1109243697478995e-06,
|
| 1425 |
+
"loss": 0.0109,
|
| 1426 |
+
"step": 10050
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 1.6974789915966386,
|
| 1430 |
+
"grad_norm": 0.002027824753895402,
|
| 1431 |
+
"learning_rate": 3.0268907563025213e-06,
|
| 1432 |
+
"loss": 0.0123,
|
| 1433 |
+
"step": 10100
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 1.7058823529411766,
|
| 1437 |
+
"grad_norm": 0.00794964749366045,
|
| 1438 |
+
"learning_rate": 2.9428571428571427e-06,
|
| 1439 |
+
"loss": 0.0066,
|
| 1440 |
+
"step": 10150
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 1.7142857142857144,
|
| 1444 |
+
"grad_norm": 7.29260778427124,
|
| 1445 |
+
"learning_rate": 2.858823529411765e-06,
|
| 1446 |
+
"loss": 0.0086,
|
| 1447 |
+
"step": 10200
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 1.7226890756302522,
|
| 1451 |
+
"grad_norm": 0.0018106248462572694,
|
| 1452 |
+
"learning_rate": 2.7747899159663867e-06,
|
| 1453 |
+
"loss": 0.0139,
|
| 1454 |
+
"step": 10250
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 1.73109243697479,
|
| 1458 |
+
"grad_norm": 17.740549087524414,
|
| 1459 |
+
"learning_rate": 2.690756302521009e-06,
|
| 1460 |
+
"loss": 0.0208,
|
| 1461 |
+
"step": 10300
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 1.7394957983193278,
|
| 1465 |
+
"grad_norm": 110.83795166015625,
|
| 1466 |
+
"learning_rate": 2.6067226890756303e-06,
|
| 1467 |
+
"loss": 0.0172,
|
| 1468 |
+
"step": 10350
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 1.7478991596638656,
|
| 1472 |
+
"grad_norm": 0.0010938843479380012,
|
| 1473 |
+
"learning_rate": 2.5226890756302525e-06,
|
| 1474 |
+
"loss": 0.0067,
|
| 1475 |
+
"step": 10400
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 1.7563025210084033,
|
| 1479 |
+
"grad_norm": 32.8217658996582,
|
| 1480 |
+
"learning_rate": 2.438655462184874e-06,
|
| 1481 |
+
"loss": 0.0202,
|
| 1482 |
+
"step": 10450
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 1.7647058823529411,
|
| 1486 |
+
"grad_norm": 0.0011374776950106025,
|
| 1487 |
+
"learning_rate": 2.3546218487394957e-06,
|
| 1488 |
+
"loss": 0.0084,
|
| 1489 |
+
"step": 10500
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 1.773109243697479,
|
| 1493 |
+
"grad_norm": 0.0036547116469591856,
|
| 1494 |
+
"learning_rate": 2.270588235294118e-06,
|
| 1495 |
+
"loss": 0.0116,
|
| 1496 |
+
"step": 10550
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 1.7815126050420167,
|
| 1500 |
+
"grad_norm": 0.11225389689207077,
|
| 1501 |
+
"learning_rate": 2.1865546218487397e-06,
|
| 1502 |
+
"loss": 0.0234,
|
| 1503 |
+
"step": 10600
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 1.7899159663865545,
|
| 1507 |
+
"grad_norm": 0.0020811427384614944,
|
| 1508 |
+
"learning_rate": 2.1025210084033615e-06,
|
| 1509 |
+
"loss": 0.0142,
|
| 1510 |
+
"step": 10650
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 1.7983193277310925,
|
| 1514 |
+
"grad_norm": 0.0006816519307903945,
|
| 1515 |
+
"learning_rate": 2.0184873949579833e-06,
|
| 1516 |
+
"loss": 0.0149,
|
| 1517 |
+
"step": 10700
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 1.8067226890756303,
|
| 1521 |
+
"grad_norm": 0.0016011211555451155,
|
| 1522 |
+
"learning_rate": 1.934453781512605e-06,
|
| 1523 |
+
"loss": 0.0218,
|
| 1524 |
+
"step": 10750
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 1.815126050420168,
|
| 1528 |
+
"grad_norm": 0.08378565311431885,
|
| 1529 |
+
"learning_rate": 1.850420168067227e-06,
|
| 1530 |
+
"loss": 0.0182,
|
| 1531 |
+
"step": 10800
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 1.8235294117647058,
|
| 1535 |
+
"grad_norm": 0.0006500572781078517,
|
| 1536 |
+
"learning_rate": 1.7663865546218489e-06,
|
| 1537 |
+
"loss": 0.0045,
|
| 1538 |
+
"step": 10850
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 1.8319327731092439,
|
| 1542 |
+
"grad_norm": 71.94564056396484,
|
| 1543 |
+
"learning_rate": 1.682352941176471e-06,
|
| 1544 |
+
"loss": 0.0005,
|
| 1545 |
+
"step": 10900
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 1.8403361344537816,
|
| 1549 |
+
"grad_norm": 0.0006799351540394127,
|
| 1550 |
+
"learning_rate": 1.5983193277310927e-06,
|
| 1551 |
+
"loss": 0.0064,
|
| 1552 |
+
"step": 10950
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 1.8487394957983194,
|
| 1556 |
+
"grad_norm": 0.9455434083938599,
|
| 1557 |
+
"learning_rate": 1.5142857142857145e-06,
|
| 1558 |
+
"loss": 0.0157,
|
| 1559 |
+
"step": 11000
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 1.8571428571428572,
|
| 1563 |
+
"grad_norm": 0.0075612375512719154,
|
| 1564 |
+
"learning_rate": 1.4302521008403363e-06,
|
| 1565 |
+
"loss": 0.0055,
|
| 1566 |
+
"step": 11050
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 1.865546218487395,
|
| 1570 |
+
"grad_norm": 0.002757304348051548,
|
| 1571 |
+
"learning_rate": 1.3462184873949579e-06,
|
| 1572 |
+
"loss": 0.0107,
|
| 1573 |
+
"step": 11100
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 1.8739495798319328,
|
| 1577 |
+
"grad_norm": 0.005417112726718187,
|
| 1578 |
+
"learning_rate": 1.2621848739495799e-06,
|
| 1579 |
+
"loss": 0.0121,
|
| 1580 |
+
"step": 11150
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 1.8823529411764706,
|
| 1584 |
+
"grad_norm": 0.0006217072950676084,
|
| 1585 |
+
"learning_rate": 1.1781512605042019e-06,
|
| 1586 |
+
"loss": 0.0105,
|
| 1587 |
+
"step": 11200
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 1.8907563025210083,
|
| 1591 |
+
"grad_norm": 0.0012244200333952904,
|
| 1592 |
+
"learning_rate": 1.0941176470588235e-06,
|
| 1593 |
+
"loss": 0.0051,
|
| 1594 |
+
"step": 11250
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 1.8991596638655461,
|
| 1598 |
+
"grad_norm": 81.42713928222656,
|
| 1599 |
+
"learning_rate": 1.0100840336134455e-06,
|
| 1600 |
+
"loss": 0.0124,
|
| 1601 |
+
"step": 11300
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 1.907563025210084,
|
| 1605 |
+
"grad_norm": 0.003745024325326085,
|
| 1606 |
+
"learning_rate": 9.260504201680673e-07,
|
| 1607 |
+
"loss": 0.0162,
|
| 1608 |
+
"step": 11350
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 1.9159663865546217,
|
| 1612 |
+
"grad_norm": 0.0008190579828806221,
|
| 1613 |
+
"learning_rate": 8.420168067226892e-07,
|
| 1614 |
+
"loss": 0.0183,
|
| 1615 |
+
"step": 11400
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 1.9243697478991597,
|
| 1619 |
+
"grad_norm": 0.0016843158518895507,
|
| 1620 |
+
"learning_rate": 7.57983193277311e-07,
|
| 1621 |
+
"loss": 0.0029,
|
| 1622 |
+
"step": 11450
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 1.9327731092436975,
|
| 1626 |
+
"grad_norm": 0.001935823354870081,
|
| 1627 |
+
"learning_rate": 6.739495798319329e-07,
|
| 1628 |
+
"loss": 0.0045,
|
| 1629 |
+
"step": 11500
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 1.9411764705882353,
|
| 1633 |
+
"grad_norm": 0.02681451104581356,
|
| 1634 |
+
"learning_rate": 5.899159663865547e-07,
|
| 1635 |
+
"loss": 0.0062,
|
| 1636 |
+
"step": 11550
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 1.949579831932773,
|
| 1640 |
+
"grad_norm": 22.993127822875977,
|
| 1641 |
+
"learning_rate": 5.058823529411766e-07,
|
| 1642 |
+
"loss": 0.0093,
|
| 1643 |
+
"step": 11600
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 1.957983193277311,
|
| 1647 |
+
"grad_norm": 0.006046156398952007,
|
| 1648 |
+
"learning_rate": 4.218487394957983e-07,
|
| 1649 |
+
"loss": 0.0094,
|
| 1650 |
+
"step": 11650
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 1.9663865546218489,
|
| 1654 |
+
"grad_norm": 0.009504588320851326,
|
| 1655 |
+
"learning_rate": 3.378151260504202e-07,
|
| 1656 |
+
"loss": 0.0108,
|
| 1657 |
+
"step": 11700
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 1.9747899159663866,
|
| 1661 |
+
"grad_norm": 9.602606773376465,
|
| 1662 |
+
"learning_rate": 2.5378151260504206e-07,
|
| 1663 |
+
"loss": 0.0034,
|
| 1664 |
+
"step": 11750
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 1.9831932773109244,
|
| 1668 |
+
"grad_norm": 0.0006558377644978464,
|
| 1669 |
+
"learning_rate": 1.6974789915966388e-07,
|
| 1670 |
+
"loss": 0.0122,
|
| 1671 |
+
"step": 11800
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 1.9915966386554622,
|
| 1675 |
+
"grad_norm": 0.0030038042459636927,
|
| 1676 |
+
"learning_rate": 8.571428571428573e-08,
|
| 1677 |
+
"loss": 0.0067,
|
| 1678 |
+
"step": 11850
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 2.0,
|
| 1682 |
+
"grad_norm": 0.0026751705445349216,
|
| 1683 |
+
"learning_rate": 1.6806722689075632e-09,
|
| 1684 |
+
"loss": 0.0039,
|
| 1685 |
+
"step": 11900
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 2.0,
|
| 1689 |
+
"eval_f1": 0.7922125924560002,
|
| 1690 |
+
"eval_loss": 1.6985238790512085,
|
| 1691 |
+
"eval_runtime": 0.0932,
|
| 1692 |
+
"eval_samples_per_second": 2340.062,
|
| 1693 |
+
"eval_steps_per_second": 42.937,
|
| 1694 |
+
"step": 11900
|
| 1695 |
+
}
|
| 1696 |
+
],
|
| 1697 |
+
"logging_steps": 50,
|
| 1698 |
+
"max_steps": 11900,
|
| 1699 |
+
"num_input_tokens_seen": 0,
|
| 1700 |
+
"num_train_epochs": 2,
|
| 1701 |
+
"save_steps": 500,
|
| 1702 |
+
"stateful_callbacks": {
|
| 1703 |
+
"TrainerControl": {
|
| 1704 |
+
"args": {
|
| 1705 |
+
"should_epoch_stop": false,
|
| 1706 |
+
"should_evaluate": false,
|
| 1707 |
+
"should_log": false,
|
| 1708 |
+
"should_save": true,
|
| 1709 |
+
"should_training_stop": true
|
| 1710 |
+
},
|
| 1711 |
+
"attributes": {}
|
| 1712 |
+
}
|
| 1713 |
+
},
|
| 1714 |
+
"total_flos": 1.6979294837936154e+16,
|
| 1715 |
+
"train_batch_size": 32,
|
| 1716 |
+
"trial_name": null,
|
| 1717 |
+
"trial_params": null
|
| 1718 |
+
}
|
checkpoint-11900/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6278d4d9aa9642cffe960097c4e2061d5852d907174ff69620fd0f70d7384ace
|
| 3 |
+
size 5841
|
checkpoint-11900/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-5950/added_tokens.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-5950/config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"directionality": "bidi",
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0",
|
| 14 |
+
"1": "LABEL_1",
|
| 15 |
+
"2": "LABEL_2",
|
| 16 |
+
"3": "LABEL_3",
|
| 17 |
+
"4": "LABEL_4"
|
| 18 |
+
},
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 3072,
|
| 21 |
+
"label2id": {
|
| 22 |
+
"LABEL_0": 0,
|
| 23 |
+
"LABEL_1": 1,
|
| 24 |
+
"LABEL_2": 2,
|
| 25 |
+
"LABEL_3": 3,
|
| 26 |
+
"LABEL_4": 4
|
| 27 |
+
},
|
| 28 |
+
"layer_norm_eps": 1e-12,
|
| 29 |
+
"max_position_embeddings": 512,
|
| 30 |
+
"model_type": "bert",
|
| 31 |
+
"num_attention_heads": 12,
|
| 32 |
+
"num_hidden_layers": 12,
|
| 33 |
+
"pad_token_id": 0,
|
| 34 |
+
"pooler_fc_size": 768,
|
| 35 |
+
"pooler_num_attention_heads": 12,
|
| 36 |
+
"pooler_num_fc_layers": 3,
|
| 37 |
+
"pooler_size_per_head": 128,
|
| 38 |
+
"pooler_type": "first_token_transform",
|
| 39 |
+
"position_embedding_type": "absolute",
|
| 40 |
+
"problem_type": "single_label_classification",
|
| 41 |
+
"transformers_version": "4.57.3",
|
| 42 |
+
"type_vocab_size": 2,
|
| 43 |
+
"use_cache": true,
|
| 44 |
+
"vocab_size": 135776
|
| 45 |
+
}
|
checkpoint-5950/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb2910d521a8ea60d31f0f00aed6ca04e89edfa7357dcb0ea416e4bced273f53
|
| 3 |
+
size 761308172
|
checkpoint-5950/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af7692ceb9fefccb3d6fd92f8297ad4a86913bdf561d2e1d9a9322643a642f26
|
| 3 |
+
size 1522740619
|
checkpoint-5950/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a49e7f281cce6213254356e0a8581c5e1a33aa8a0c4f3d82358a49c0dcad01e
|
| 3 |
+
size 14645
|
checkpoint-5950/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60a4a8f30c523d62fefef6234caad68c149fe116459f430c36b44d72837c6777
|
| 3 |
+
size 1383
|
checkpoint-5950/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b9393c5db773800350e887b5a2d258030bcab33026356e5a0cc7b53c6b6e904c
|
| 3 |
+
size 1465
|
checkpoint-5950/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
checkpoint-5950/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-5950/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-5950/trainer_state.json
ADDED
|
@@ -0,0 +1,876 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 5950,
|
| 3 |
+
"best_metric": 0.7729265244559362,
|
| 4 |
+
"best_model_checkpoint": "pseudo_model/checkpoint-5950",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 5950,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008403361344537815,
|
| 14 |
+
"grad_norm": 70.3099365234375,
|
| 15 |
+
"learning_rate": 1.991764705882353e-05,
|
| 16 |
+
"loss": 0.0639,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01680672268907563,
|
| 21 |
+
"grad_norm": 0.251952201128006,
|
| 22 |
+
"learning_rate": 1.983361344537815e-05,
|
| 23 |
+
"loss": 0.1373,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.025210084033613446,
|
| 28 |
+
"grad_norm": 54.500118255615234,
|
| 29 |
+
"learning_rate": 1.9749579831932776e-05,
|
| 30 |
+
"loss": 0.187,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03361344537815126,
|
| 35 |
+
"grad_norm": 17.24431037902832,
|
| 36 |
+
"learning_rate": 1.9665546218487398e-05,
|
| 37 |
+
"loss": 0.1268,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04201680672268908,
|
| 42 |
+
"grad_norm": 0.7904719114303589,
|
| 43 |
+
"learning_rate": 1.958151260504202e-05,
|
| 44 |
+
"loss": 0.1147,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05042016806722689,
|
| 49 |
+
"grad_norm": 15.803120613098145,
|
| 50 |
+
"learning_rate": 1.949747899159664e-05,
|
| 51 |
+
"loss": 0.0939,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.058823529411764705,
|
| 56 |
+
"grad_norm": 21.106233596801758,
|
| 57 |
+
"learning_rate": 1.9413445378151262e-05,
|
| 58 |
+
"loss": 0.1392,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06722689075630252,
|
| 63 |
+
"grad_norm": 4.552457809448242,
|
| 64 |
+
"learning_rate": 1.9329411764705883e-05,
|
| 65 |
+
"loss": 0.0754,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07563025210084033,
|
| 70 |
+
"grad_norm": 59.192100524902344,
|
| 71 |
+
"learning_rate": 1.9245378151260505e-05,
|
| 72 |
+
"loss": 0.0864,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08403361344537816,
|
| 77 |
+
"grad_norm": 24.423683166503906,
|
| 78 |
+
"learning_rate": 1.9161344537815126e-05,
|
| 79 |
+
"loss": 0.1462,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09243697478991597,
|
| 84 |
+
"grad_norm": 20.817564010620117,
|
| 85 |
+
"learning_rate": 1.907731092436975e-05,
|
| 86 |
+
"loss": 0.1025,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10084033613445378,
|
| 91 |
+
"grad_norm": 16.254493713378906,
|
| 92 |
+
"learning_rate": 1.8993277310924372e-05,
|
| 93 |
+
"loss": 0.1588,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1092436974789916,
|
| 98 |
+
"grad_norm": 43.414573669433594,
|
| 99 |
+
"learning_rate": 1.8909243697478993e-05,
|
| 100 |
+
"loss": 0.0737,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.11764705882352941,
|
| 105 |
+
"grad_norm": 25.70012664794922,
|
| 106 |
+
"learning_rate": 1.8825210084033615e-05,
|
| 107 |
+
"loss": 0.0638,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.12605042016806722,
|
| 112 |
+
"grad_norm": 4.720924377441406,
|
| 113 |
+
"learning_rate": 1.8741176470588236e-05,
|
| 114 |
+
"loss": 0.1043,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.13445378151260504,
|
| 119 |
+
"grad_norm": 57.87378692626953,
|
| 120 |
+
"learning_rate": 1.8657142857142858e-05,
|
| 121 |
+
"loss": 0.1188,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.14285714285714285,
|
| 126 |
+
"grad_norm": 23.307783126831055,
|
| 127 |
+
"learning_rate": 1.857310924369748e-05,
|
| 128 |
+
"loss": 0.1161,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.15126050420168066,
|
| 133 |
+
"grad_norm": 44.87445831298828,
|
| 134 |
+
"learning_rate": 1.8489075630252104e-05,
|
| 135 |
+
"loss": 0.0917,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.15966386554621848,
|
| 140 |
+
"grad_norm": 14.766839981079102,
|
| 141 |
+
"learning_rate": 1.8405042016806725e-05,
|
| 142 |
+
"loss": 0.1296,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.16806722689075632,
|
| 147 |
+
"grad_norm": 8.999948501586914,
|
| 148 |
+
"learning_rate": 1.8321008403361346e-05,
|
| 149 |
+
"loss": 0.0966,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.17647058823529413,
|
| 154 |
+
"grad_norm": 27.70500946044922,
|
| 155 |
+
"learning_rate": 1.8236974789915968e-05,
|
| 156 |
+
"loss": 0.0847,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.18487394957983194,
|
| 161 |
+
"grad_norm": 45.68607711791992,
|
| 162 |
+
"learning_rate": 1.815294117647059e-05,
|
| 163 |
+
"loss": 0.0774,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.19327731092436976,
|
| 168 |
+
"grad_norm": 9.21766185760498,
|
| 169 |
+
"learning_rate": 1.806890756302521e-05,
|
| 170 |
+
"loss": 0.0832,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.20168067226890757,
|
| 175 |
+
"grad_norm": 84.41997528076172,
|
| 176 |
+
"learning_rate": 1.7984873949579832e-05,
|
| 177 |
+
"loss": 0.1061,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.21008403361344538,
|
| 182 |
+
"grad_norm": 11.963436126708984,
|
| 183 |
+
"learning_rate": 1.7900840336134457e-05,
|
| 184 |
+
"loss": 0.1119,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2184873949579832,
|
| 189 |
+
"grad_norm": 30.294477462768555,
|
| 190 |
+
"learning_rate": 1.7816806722689078e-05,
|
| 191 |
+
"loss": 0.089,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.226890756302521,
|
| 196 |
+
"grad_norm": 0.17758768796920776,
|
| 197 |
+
"learning_rate": 1.77327731092437e-05,
|
| 198 |
+
"loss": 0.0671,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.23529411764705882,
|
| 203 |
+
"grad_norm": 0.24741335213184357,
|
| 204 |
+
"learning_rate": 1.764873949579832e-05,
|
| 205 |
+
"loss": 0.0875,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.24369747899159663,
|
| 210 |
+
"grad_norm": 79.4989013671875,
|
| 211 |
+
"learning_rate": 1.7564705882352942e-05,
|
| 212 |
+
"loss": 0.139,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.25210084033613445,
|
| 217 |
+
"grad_norm": 40.87698745727539,
|
| 218 |
+
"learning_rate": 1.7480672268907564e-05,
|
| 219 |
+
"loss": 0.0666,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2605042016806723,
|
| 224 |
+
"grad_norm": 4.265988826751709,
|
| 225 |
+
"learning_rate": 1.7396638655462185e-05,
|
| 226 |
+
"loss": 0.0957,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2689075630252101,
|
| 231 |
+
"grad_norm": 31.657917022705078,
|
| 232 |
+
"learning_rate": 1.7312605042016806e-05,
|
| 233 |
+
"loss": 0.1013,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2773109243697479,
|
| 238 |
+
"grad_norm": 48.78037643432617,
|
| 239 |
+
"learning_rate": 1.722857142857143e-05,
|
| 240 |
+
"loss": 0.0721,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2857142857142857,
|
| 245 |
+
"grad_norm": 0.1120712086558342,
|
| 246 |
+
"learning_rate": 1.7144537815126052e-05,
|
| 247 |
+
"loss": 0.0865,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.29411764705882354,
|
| 252 |
+
"grad_norm": 0.8699374198913574,
|
| 253 |
+
"learning_rate": 1.7060504201680674e-05,
|
| 254 |
+
"loss": 0.0779,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3025210084033613,
|
| 259 |
+
"grad_norm": 64.08531951904297,
|
| 260 |
+
"learning_rate": 1.6976470588235295e-05,
|
| 261 |
+
"loss": 0.0938,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.31092436974789917,
|
| 266 |
+
"grad_norm": 36.72616195678711,
|
| 267 |
+
"learning_rate": 1.6892436974789917e-05,
|
| 268 |
+
"loss": 0.068,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.31932773109243695,
|
| 273 |
+
"grad_norm": 24.38673973083496,
|
| 274 |
+
"learning_rate": 1.6808403361344538e-05,
|
| 275 |
+
"loss": 0.0667,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3277310924369748,
|
| 280 |
+
"grad_norm": 4.634633541107178,
|
| 281 |
+
"learning_rate": 1.672436974789916e-05,
|
| 282 |
+
"loss": 0.1098,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.33613445378151263,
|
| 287 |
+
"grad_norm": 0.005323654506355524,
|
| 288 |
+
"learning_rate": 1.6640336134453784e-05,
|
| 289 |
+
"loss": 0.0578,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3445378151260504,
|
| 294 |
+
"grad_norm": 30.870010375976562,
|
| 295 |
+
"learning_rate": 1.6556302521008405e-05,
|
| 296 |
+
"loss": 0.0702,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.35294117647058826,
|
| 301 |
+
"grad_norm": 0.3597383201122284,
|
| 302 |
+
"learning_rate": 1.6472268907563027e-05,
|
| 303 |
+
"loss": 0.087,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.36134453781512604,
|
| 308 |
+
"grad_norm": 43.77628707885742,
|
| 309 |
+
"learning_rate": 1.6388235294117648e-05,
|
| 310 |
+
"loss": 0.0865,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3697478991596639,
|
| 315 |
+
"grad_norm": 5.36348819732666,
|
| 316 |
+
"learning_rate": 1.630420168067227e-05,
|
| 317 |
+
"loss": 0.1211,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.37815126050420167,
|
| 322 |
+
"grad_norm": 5.006982326507568,
|
| 323 |
+
"learning_rate": 1.622016806722689e-05,
|
| 324 |
+
"loss": 0.0772,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3865546218487395,
|
| 329 |
+
"grad_norm": 0.21585534512996674,
|
| 330 |
+
"learning_rate": 1.6136134453781512e-05,
|
| 331 |
+
"loss": 0.0873,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3949579831932773,
|
| 336 |
+
"grad_norm": 5.9591193199157715,
|
| 337 |
+
"learning_rate": 1.6052100840336137e-05,
|
| 338 |
+
"loss": 0.0678,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.40336134453781514,
|
| 343 |
+
"grad_norm": 5.950344085693359,
|
| 344 |
+
"learning_rate": 1.596806722689076e-05,
|
| 345 |
+
"loss": 0.0583,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.4117647058823529,
|
| 350 |
+
"grad_norm": 62.205570220947266,
|
| 351 |
+
"learning_rate": 1.588403361344538e-05,
|
| 352 |
+
"loss": 0.0799,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.42016806722689076,
|
| 357 |
+
"grad_norm": 11.344050407409668,
|
| 358 |
+
"learning_rate": 1.58e-05,
|
| 359 |
+
"loss": 0.0925,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.42857142857142855,
|
| 364 |
+
"grad_norm": 20.887693405151367,
|
| 365 |
+
"learning_rate": 1.5715966386554623e-05,
|
| 366 |
+
"loss": 0.0515,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.4369747899159664,
|
| 371 |
+
"grad_norm": 4.148672580718994,
|
| 372 |
+
"learning_rate": 1.5631932773109244e-05,
|
| 373 |
+
"loss": 0.0836,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.44537815126050423,
|
| 378 |
+
"grad_norm": 0.019104987382888794,
|
| 379 |
+
"learning_rate": 1.5547899159663865e-05,
|
| 380 |
+
"loss": 0.0663,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.453781512605042,
|
| 385 |
+
"grad_norm": 55.08736038208008,
|
| 386 |
+
"learning_rate": 1.5463865546218487e-05,
|
| 387 |
+
"loss": 0.0573,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.46218487394957986,
|
| 392 |
+
"grad_norm": 24.77686882019043,
|
| 393 |
+
"learning_rate": 1.537983193277311e-05,
|
| 394 |
+
"loss": 0.0995,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.47058823529411764,
|
| 399 |
+
"grad_norm": 0.2827483117580414,
|
| 400 |
+
"learning_rate": 1.5295798319327733e-05,
|
| 401 |
+
"loss": 0.0931,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4789915966386555,
|
| 406 |
+
"grad_norm": 51.127052307128906,
|
| 407 |
+
"learning_rate": 1.5211764705882354e-05,
|
| 408 |
+
"loss": 0.0674,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.48739495798319327,
|
| 413 |
+
"grad_norm": 28.012380599975586,
|
| 414 |
+
"learning_rate": 1.5127731092436977e-05,
|
| 415 |
+
"loss": 0.071,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.4957983193277311,
|
| 420 |
+
"grad_norm": 62.96744155883789,
|
| 421 |
+
"learning_rate": 1.5043697478991597e-05,
|
| 422 |
+
"loss": 0.0802,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5042016806722689,
|
| 427 |
+
"grad_norm": 24.443973541259766,
|
| 428 |
+
"learning_rate": 1.4959663865546218e-05,
|
| 429 |
+
"loss": 0.0822,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5126050420168067,
|
| 434 |
+
"grad_norm": 0.09382598847150803,
|
| 435 |
+
"learning_rate": 1.4875630252100841e-05,
|
| 436 |
+
"loss": 0.0893,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5210084033613446,
|
| 441 |
+
"grad_norm": 8.969447135925293,
|
| 442 |
+
"learning_rate": 1.4791596638655463e-05,
|
| 443 |
+
"loss": 0.0587,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5294117647058824,
|
| 448 |
+
"grad_norm": 0.33123093843460083,
|
| 449 |
+
"learning_rate": 1.4707563025210086e-05,
|
| 450 |
+
"loss": 0.0662,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5378151260504201,
|
| 455 |
+
"grad_norm": 0.34799447655677795,
|
| 456 |
+
"learning_rate": 1.4623529411764707e-05,
|
| 457 |
+
"loss": 0.0471,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5462184873949579,
|
| 462 |
+
"grad_norm": 0.10502827167510986,
|
| 463 |
+
"learning_rate": 1.453949579831933e-05,
|
| 464 |
+
"loss": 0.0608,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5546218487394958,
|
| 469 |
+
"grad_norm": 46.54685974121094,
|
| 470 |
+
"learning_rate": 1.4455462184873952e-05,
|
| 471 |
+
"loss": 0.0756,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5630252100840336,
|
| 476 |
+
"grad_norm": 25.307750701904297,
|
| 477 |
+
"learning_rate": 1.4371428571428571e-05,
|
| 478 |
+
"loss": 0.0703,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5714285714285714,
|
| 483 |
+
"grad_norm": 0.03593587502837181,
|
| 484 |
+
"learning_rate": 1.4287394957983194e-05,
|
| 485 |
+
"loss": 0.0509,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5798319327731093,
|
| 490 |
+
"grad_norm": 6.873576641082764,
|
| 491 |
+
"learning_rate": 1.4203361344537816e-05,
|
| 492 |
+
"loss": 0.0751,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5882352941176471,
|
| 497 |
+
"grad_norm": 39.44701385498047,
|
| 498 |
+
"learning_rate": 1.4119327731092439e-05,
|
| 499 |
+
"loss": 0.0698,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5966386554621849,
|
| 504 |
+
"grad_norm": 0.7871583104133606,
|
| 505 |
+
"learning_rate": 1.403529411764706e-05,
|
| 506 |
+
"loss": 0.0701,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6050420168067226,
|
| 511 |
+
"grad_norm": 0.06720981746912003,
|
| 512 |
+
"learning_rate": 1.3951260504201683e-05,
|
| 513 |
+
"loss": 0.0613,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6134453781512605,
|
| 518 |
+
"grad_norm": 87.64744567871094,
|
| 519 |
+
"learning_rate": 1.3867226890756305e-05,
|
| 520 |
+
"loss": 0.0657,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6218487394957983,
|
| 525 |
+
"grad_norm": 92.26810455322266,
|
| 526 |
+
"learning_rate": 1.3783193277310924e-05,
|
| 527 |
+
"loss": 0.0915,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6302521008403361,
|
| 532 |
+
"grad_norm": 0.13392935693264008,
|
| 533 |
+
"learning_rate": 1.3699159663865547e-05,
|
| 534 |
+
"loss": 0.0552,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6386554621848739,
|
| 539 |
+
"grad_norm": 0.2624678313732147,
|
| 540 |
+
"learning_rate": 1.3615126050420169e-05,
|
| 541 |
+
"loss": 0.061,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6470588235294118,
|
| 546 |
+
"grad_norm": 58.510833740234375,
|
| 547 |
+
"learning_rate": 1.353109243697479e-05,
|
| 548 |
+
"loss": 0.0475,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6554621848739496,
|
| 553 |
+
"grad_norm": 0.19354496896266937,
|
| 554 |
+
"learning_rate": 1.3447058823529413e-05,
|
| 555 |
+
"loss": 0.0423,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6638655462184874,
|
| 560 |
+
"grad_norm": 20.584426879882812,
|
| 561 |
+
"learning_rate": 1.3363025210084035e-05,
|
| 562 |
+
"loss": 0.1135,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6722689075630253,
|
| 567 |
+
"grad_norm": 1.4565203189849854,
|
| 568 |
+
"learning_rate": 1.3278991596638658e-05,
|
| 569 |
+
"loss": 0.0264,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.680672268907563,
|
| 574 |
+
"grad_norm": 0.0815153419971466,
|
| 575 |
+
"learning_rate": 1.3194957983193279e-05,
|
| 576 |
+
"loss": 0.0956,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6890756302521008,
|
| 581 |
+
"grad_norm": 45.101993560791016,
|
| 582 |
+
"learning_rate": 1.3110924369747899e-05,
|
| 583 |
+
"loss": 0.0853,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6974789915966386,
|
| 588 |
+
"grad_norm": 8.502076148986816,
|
| 589 |
+
"learning_rate": 1.3026890756302522e-05,
|
| 590 |
+
"loss": 0.0407,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7058823529411765,
|
| 595 |
+
"grad_norm": 0.01488969475030899,
|
| 596 |
+
"learning_rate": 1.2942857142857143e-05,
|
| 597 |
+
"loss": 0.0506,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7142857142857143,
|
| 602 |
+
"grad_norm": 128.31639099121094,
|
| 603 |
+
"learning_rate": 1.2858823529411766e-05,
|
| 604 |
+
"loss": 0.0527,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.7226890756302521,
|
| 609 |
+
"grad_norm": 69.5967025756836,
|
| 610 |
+
"learning_rate": 1.2774789915966388e-05,
|
| 611 |
+
"loss": 0.0437,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7310924369747899,
|
| 616 |
+
"grad_norm": 51.27083206176758,
|
| 617 |
+
"learning_rate": 1.269075630252101e-05,
|
| 618 |
+
"loss": 0.0511,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7394957983193278,
|
| 623 |
+
"grad_norm": 1.9427380561828613,
|
| 624 |
+
"learning_rate": 1.2606722689075632e-05,
|
| 625 |
+
"loss": 0.0938,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7478991596638656,
|
| 630 |
+
"grad_norm": 17.49405288696289,
|
| 631 |
+
"learning_rate": 1.2522689075630252e-05,
|
| 632 |
+
"loss": 0.0799,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7563025210084033,
|
| 637 |
+
"grad_norm": 2.69028377532959,
|
| 638 |
+
"learning_rate": 1.2438655462184875e-05,
|
| 639 |
+
"loss": 0.0556,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7647058823529411,
|
| 644 |
+
"grad_norm": 91.98591613769531,
|
| 645 |
+
"learning_rate": 1.2354621848739496e-05,
|
| 646 |
+
"loss": 0.0341,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.773109243697479,
|
| 651 |
+
"grad_norm": 37.63540267944336,
|
| 652 |
+
"learning_rate": 1.2270588235294119e-05,
|
| 653 |
+
"loss": 0.0748,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7815126050420168,
|
| 658 |
+
"grad_norm": 24.502737045288086,
|
| 659 |
+
"learning_rate": 1.218655462184874e-05,
|
| 660 |
+
"loss": 0.0693,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7899159663865546,
|
| 665 |
+
"grad_norm": 0.010893125087022781,
|
| 666 |
+
"learning_rate": 1.2102521008403364e-05,
|
| 667 |
+
"loss": 0.0371,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7983193277310925,
|
| 672 |
+
"grad_norm": 39.78943634033203,
|
| 673 |
+
"learning_rate": 1.2018487394957985e-05,
|
| 674 |
+
"loss": 0.0547,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.8067226890756303,
|
| 679 |
+
"grad_norm": 21.680444717407227,
|
| 680 |
+
"learning_rate": 1.1934453781512605e-05,
|
| 681 |
+
"loss": 0.0405,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8151260504201681,
|
| 686 |
+
"grad_norm": 91.88570404052734,
|
| 687 |
+
"learning_rate": 1.1850420168067228e-05,
|
| 688 |
+
"loss": 0.048,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8235294117647058,
|
| 693 |
+
"grad_norm": 47.21152114868164,
|
| 694 |
+
"learning_rate": 1.1766386554621849e-05,
|
| 695 |
+
"loss": 0.0957,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8319327731092437,
|
| 700 |
+
"grad_norm": 0.03324070945382118,
|
| 701 |
+
"learning_rate": 1.1682352941176472e-05,
|
| 702 |
+
"loss": 0.0525,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8403361344537815,
|
| 707 |
+
"grad_norm": 0.4629770517349243,
|
| 708 |
+
"learning_rate": 1.1598319327731094e-05,
|
| 709 |
+
"loss": 0.0682,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8487394957983193,
|
| 714 |
+
"grad_norm": 45.92827224731445,
|
| 715 |
+
"learning_rate": 1.1514285714285715e-05,
|
| 716 |
+
"loss": 0.0424,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8571428571428571,
|
| 721 |
+
"grad_norm": 32.47571563720703,
|
| 722 |
+
"learning_rate": 1.1430252100840338e-05,
|
| 723 |
+
"loss": 0.0794,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.865546218487395,
|
| 728 |
+
"grad_norm": 17.017562866210938,
|
| 729 |
+
"learning_rate": 1.134621848739496e-05,
|
| 730 |
+
"loss": 0.0436,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8739495798319328,
|
| 735 |
+
"grad_norm": 2.4690027236938477,
|
| 736 |
+
"learning_rate": 1.1262184873949579e-05,
|
| 737 |
+
"loss": 0.0618,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8823529411764706,
|
| 742 |
+
"grad_norm": 91.57916259765625,
|
| 743 |
+
"learning_rate": 1.1178151260504202e-05,
|
| 744 |
+
"loss": 0.0628,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8907563025210085,
|
| 749 |
+
"grad_norm": 0.11016395688056946,
|
| 750 |
+
"learning_rate": 1.1094117647058823e-05,
|
| 751 |
+
"loss": 0.076,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8991596638655462,
|
| 756 |
+
"grad_norm": 42.38503646850586,
|
| 757 |
+
"learning_rate": 1.1010084033613447e-05,
|
| 758 |
+
"loss": 0.0759,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.907563025210084,
|
| 763 |
+
"grad_norm": 4.813717365264893,
|
| 764 |
+
"learning_rate": 1.0926050420168068e-05,
|
| 765 |
+
"loss": 0.0349,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.9159663865546218,
|
| 770 |
+
"grad_norm": 0.016100173816084862,
|
| 771 |
+
"learning_rate": 1.0842016806722691e-05,
|
| 772 |
+
"loss": 0.0506,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.9243697478991597,
|
| 777 |
+
"grad_norm": 0.07725575566291809,
|
| 778 |
+
"learning_rate": 1.0757983193277312e-05,
|
| 779 |
+
"loss": 0.0581,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9327731092436975,
|
| 784 |
+
"grad_norm": 69.73644256591797,
|
| 785 |
+
"learning_rate": 1.0673949579831932e-05,
|
| 786 |
+
"loss": 0.0552,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9411764705882353,
|
| 791 |
+
"grad_norm": 0.0879245325922966,
|
| 792 |
+
"learning_rate": 1.0589915966386555e-05,
|
| 793 |
+
"loss": 0.059,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9495798319327731,
|
| 798 |
+
"grad_norm": 0.06376896053552628,
|
| 799 |
+
"learning_rate": 1.0505882352941176e-05,
|
| 800 |
+
"loss": 0.046,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.957983193277311,
|
| 805 |
+
"grad_norm": 98.75856018066406,
|
| 806 |
+
"learning_rate": 1.04218487394958e-05,
|
| 807 |
+
"loss": 0.0319,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9663865546218487,
|
| 812 |
+
"grad_norm": 0.07460715621709824,
|
| 813 |
+
"learning_rate": 1.0337815126050421e-05,
|
| 814 |
+
"loss": 0.0349,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9747899159663865,
|
| 819 |
+
"grad_norm": 88.64122772216797,
|
| 820 |
+
"learning_rate": 1.0253781512605044e-05,
|
| 821 |
+
"loss": 0.0577,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9831932773109243,
|
| 826 |
+
"grad_norm": 0.0315723791718483,
|
| 827 |
+
"learning_rate": 1.0169747899159665e-05,
|
| 828 |
+
"loss": 0.0655,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9915966386554622,
|
| 833 |
+
"grad_norm": 74.6689682006836,
|
| 834 |
+
"learning_rate": 1.0085714285714288e-05,
|
| 835 |
+
"loss": 0.0457,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.0,
|
| 840 |
+
"grad_norm": 0.06222504749894142,
|
| 841 |
+
"learning_rate": 1.0001680672268908e-05,
|
| 842 |
+
"loss": 0.0391,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.0,
|
| 847 |
+
"eval_f1": 0.7729265244559362,
|
| 848 |
+
"eval_loss": 1.5595977306365967,
|
| 849 |
+
"eval_runtime": 0.0922,
|
| 850 |
+
"eval_samples_per_second": 2364.65,
|
| 851 |
+
"eval_steps_per_second": 43.388,
|
| 852 |
+
"step": 5950
|
| 853 |
+
}
|
| 854 |
+
],
|
| 855 |
+
"logging_steps": 50,
|
| 856 |
+
"max_steps": 11900,
|
| 857 |
+
"num_input_tokens_seen": 0,
|
| 858 |
+
"num_train_epochs": 2,
|
| 859 |
+
"save_steps": 500,
|
| 860 |
+
"stateful_callbacks": {
|
| 861 |
+
"TrainerControl": {
|
| 862 |
+
"args": {
|
| 863 |
+
"should_epoch_stop": false,
|
| 864 |
+
"should_evaluate": false,
|
| 865 |
+
"should_log": false,
|
| 866 |
+
"should_save": true,
|
| 867 |
+
"should_training_stop": false
|
| 868 |
+
},
|
| 869 |
+
"attributes": {}
|
| 870 |
+
}
|
| 871 |
+
},
|
| 872 |
+
"total_flos": 8495403899188410.0,
|
| 873 |
+
"train_batch_size": 32,
|
| 874 |
+
"trial_name": null,
|
| 875 |
+
"trial_params": null
|
| 876 |
+
}
|
checkpoint-5950/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6278d4d9aa9642cffe960097c4e2061d5852d907174ff69620fd0f70d7384ace
|
| 3 |
+
size 5841
|
checkpoint-5950/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|