Training in progress, epoch 1
Browse files- checkpoint-131/config.json +31 -0
- checkpoint-131/optimizer.pt +3 -0
- checkpoint-131/pytorch_model.bin +3 -0
- checkpoint-131/rng_state.pth +3 -0
- checkpoint-131/scaler.pt +3 -0
- checkpoint-131/scheduler.pt +3 -0
- checkpoint-131/special_tokens_map.json +7 -0
- checkpoint-131/tokenizer.json +0 -0
- checkpoint-131/tokenizer_config.json +14 -0
- checkpoint-131/trainer_state.json +812 -0
- checkpoint-131/training_args.bin +3 -0
- checkpoint-131/vocab.txt +0 -0
- config.json +31 -0
- pytorch_model.bin +3 -0
- runs/Jun18_08-13-22_03eed8418a32/1655540012.1337984/events.out.tfevents.1655540012.03eed8418a32.67.3 +3 -0
- runs/Jun18_08-13-22_03eed8418a32/events.out.tfevents.1655540012.03eed8418a32.67.2 +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
checkpoint-131/config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "S2312dal/M4_MLM",
|
| 3 |
+
"activation": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"DistilBertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"hidden_dim": 3072,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0"
|
| 13 |
+
},
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"label2id": {
|
| 16 |
+
"LABEL_0": 0
|
| 17 |
+
},
|
| 18 |
+
"max_position_embeddings": 512,
|
| 19 |
+
"model_type": "distilbert",
|
| 20 |
+
"n_heads": 12,
|
| 21 |
+
"n_layers": 6,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"problem_type": "regression",
|
| 24 |
+
"qa_dropout": 0.1,
|
| 25 |
+
"seq_classif_dropout": 0.2,
|
| 26 |
+
"sinusoidal_pos_embds": false,
|
| 27 |
+
"tie_weights_": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.20.0",
|
| 30 |
+
"vocab_size": 30522
|
| 31 |
+
}
|
checkpoint-131/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29db7fa8a1b664a141cc0006183956ab98f88fc874a7b0f41dddef17789098bf
|
| 3 |
+
size 535693793
|
checkpoint-131/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e24a89fd74ece8f3fa590ef12e846792801810dbb15c7ac06b70db892a0250e5
|
| 3 |
+
size 267851249
|
checkpoint-131/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea5baa25319e85936a7201b7c74ed4a46755daa755a88f613c380b84956b213b
|
| 3 |
+
size 14503
|
checkpoint-131/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e8431068e2a13416902cb5eea86ac81660fed534fc56a772633b1d351968db8
|
| 3 |
+
size 559
|
checkpoint-131/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:461023d55f72648d3de5245756d385298240c3b10eda56e19b1987f6b67ed071
|
| 3 |
+
size 623
|
checkpoint-131/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-131/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-131/tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"do_lower_case": true,
|
| 4 |
+
"mask_token": "[MASK]",
|
| 5 |
+
"model_max_length": 512,
|
| 6 |
+
"name_or_path": "S2312dal/M4_MLM",
|
| 7 |
+
"pad_token": "[PAD]",
|
| 8 |
+
"sep_token": "[SEP]",
|
| 9 |
+
"special_tokens_map_file": null,
|
| 10 |
+
"strip_accents": null,
|
| 11 |
+
"tokenize_chinese_chars": true,
|
| 12 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 13 |
+
"unk_token": "[UNK]"
|
| 14 |
+
}
|
checkpoint-131/trainer_state.json
ADDED
|
@@ -0,0 +1,812 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"global_step": 131,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 2.5e-06,
|
| 13 |
+
"loss": 0.3281,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 5e-06,
|
| 19 |
+
"loss": 0.3522,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.02,
|
| 24 |
+
"learning_rate": 7.500000000000001e-06,
|
| 25 |
+
"loss": 0.2372,
|
| 26 |
+
"step": 3
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.03,
|
| 30 |
+
"learning_rate": 1e-05,
|
| 31 |
+
"loss": 0.2043,
|
| 32 |
+
"step": 4
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.04,
|
| 36 |
+
"learning_rate": 1.25e-05,
|
| 37 |
+
"loss": 0.3516,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05,
|
| 42 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 43 |
+
"loss": 0.2391,
|
| 44 |
+
"step": 6
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05,
|
| 48 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 49 |
+
"loss": 0.1406,
|
| 50 |
+
"step": 7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.06,
|
| 54 |
+
"learning_rate": 2e-05,
|
| 55 |
+
"loss": 0.1592,
|
| 56 |
+
"step": 8
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.07,
|
| 60 |
+
"learning_rate": 1.994805194805195e-05,
|
| 61 |
+
"loss": 0.2057,
|
| 62 |
+
"step": 9
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.08,
|
| 66 |
+
"learning_rate": 1.98961038961039e-05,
|
| 67 |
+
"loss": 0.2209,
|
| 68 |
+
"step": 10
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.08,
|
| 72 |
+
"learning_rate": 1.9844155844155846e-05,
|
| 73 |
+
"loss": 0.2098,
|
| 74 |
+
"step": 11
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.09,
|
| 78 |
+
"learning_rate": 1.9792207792207794e-05,
|
| 79 |
+
"loss": 0.2233,
|
| 80 |
+
"step": 12
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1,
|
| 84 |
+
"learning_rate": 1.974025974025974e-05,
|
| 85 |
+
"loss": 0.1877,
|
| 86 |
+
"step": 13
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.11,
|
| 90 |
+
"learning_rate": 1.968831168831169e-05,
|
| 91 |
+
"loss": 0.2094,
|
| 92 |
+
"step": 14
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.11,
|
| 96 |
+
"learning_rate": 1.963636363636364e-05,
|
| 97 |
+
"loss": 0.1608,
|
| 98 |
+
"step": 15
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.12,
|
| 102 |
+
"learning_rate": 1.9584415584415586e-05,
|
| 103 |
+
"loss": 0.1873,
|
| 104 |
+
"step": 16
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.13,
|
| 108 |
+
"learning_rate": 1.9532467532467533e-05,
|
| 109 |
+
"loss": 0.2251,
|
| 110 |
+
"step": 17
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.14,
|
| 114 |
+
"learning_rate": 1.9480519480519483e-05,
|
| 115 |
+
"loss": 0.2143,
|
| 116 |
+
"step": 18
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.15,
|
| 120 |
+
"learning_rate": 1.942857142857143e-05,
|
| 121 |
+
"loss": 0.179,
|
| 122 |
+
"step": 19
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.15,
|
| 126 |
+
"learning_rate": 1.9376623376623377e-05,
|
| 127 |
+
"loss": 0.2063,
|
| 128 |
+
"step": 20
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.16,
|
| 132 |
+
"learning_rate": 1.9324675324675325e-05,
|
| 133 |
+
"loss": 0.1789,
|
| 134 |
+
"step": 21
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.17,
|
| 138 |
+
"learning_rate": 1.9272727272727275e-05,
|
| 139 |
+
"loss": 0.1918,
|
| 140 |
+
"step": 22
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.18,
|
| 144 |
+
"learning_rate": 1.9220779220779222e-05,
|
| 145 |
+
"loss": 0.1587,
|
| 146 |
+
"step": 23
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.18,
|
| 150 |
+
"learning_rate": 1.916883116883117e-05,
|
| 151 |
+
"loss": 0.1268,
|
| 152 |
+
"step": 24
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.19,
|
| 156 |
+
"learning_rate": 1.9116883116883117e-05,
|
| 157 |
+
"loss": 0.1884,
|
| 158 |
+
"step": 25
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.2,
|
| 162 |
+
"learning_rate": 1.9064935064935067e-05,
|
| 163 |
+
"loss": 0.2015,
|
| 164 |
+
"step": 26
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.21,
|
| 168 |
+
"learning_rate": 1.9012987012987014e-05,
|
| 169 |
+
"loss": 0.1141,
|
| 170 |
+
"step": 27
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.21,
|
| 174 |
+
"learning_rate": 1.896103896103896e-05,
|
| 175 |
+
"loss": 0.2242,
|
| 176 |
+
"step": 28
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.22,
|
| 180 |
+
"learning_rate": 1.8909090909090912e-05,
|
| 181 |
+
"loss": 0.1462,
|
| 182 |
+
"step": 29
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.23,
|
| 186 |
+
"learning_rate": 1.885714285714286e-05,
|
| 187 |
+
"loss": 0.1986,
|
| 188 |
+
"step": 30
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.24,
|
| 192 |
+
"learning_rate": 1.8805194805194806e-05,
|
| 193 |
+
"loss": 0.1237,
|
| 194 |
+
"step": 31
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.24,
|
| 198 |
+
"learning_rate": 1.8753246753246753e-05,
|
| 199 |
+
"loss": 0.1657,
|
| 200 |
+
"step": 32
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.25,
|
| 204 |
+
"learning_rate": 1.8701298701298704e-05,
|
| 205 |
+
"loss": 0.1336,
|
| 206 |
+
"step": 33
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.26,
|
| 210 |
+
"learning_rate": 1.864935064935065e-05,
|
| 211 |
+
"loss": 0.1469,
|
| 212 |
+
"step": 34
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.27,
|
| 216 |
+
"learning_rate": 1.8597402597402598e-05,
|
| 217 |
+
"loss": 0.1352,
|
| 218 |
+
"step": 35
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.27,
|
| 222 |
+
"learning_rate": 1.8545454545454545e-05,
|
| 223 |
+
"loss": 0.1617,
|
| 224 |
+
"step": 36
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.28,
|
| 228 |
+
"learning_rate": 1.8493506493506496e-05,
|
| 229 |
+
"loss": 0.1403,
|
| 230 |
+
"step": 37
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.29,
|
| 234 |
+
"learning_rate": 1.8441558441558443e-05,
|
| 235 |
+
"loss": 0.1255,
|
| 236 |
+
"step": 38
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.3,
|
| 240 |
+
"learning_rate": 1.838961038961039e-05,
|
| 241 |
+
"loss": 0.102,
|
| 242 |
+
"step": 39
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.31,
|
| 246 |
+
"learning_rate": 1.8337662337662337e-05,
|
| 247 |
+
"loss": 0.1318,
|
| 248 |
+
"step": 40
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.31,
|
| 252 |
+
"learning_rate": 1.8285714285714288e-05,
|
| 253 |
+
"loss": 0.117,
|
| 254 |
+
"step": 41
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.32,
|
| 258 |
+
"learning_rate": 1.8233766233766235e-05,
|
| 259 |
+
"loss": 0.0931,
|
| 260 |
+
"step": 42
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.33,
|
| 264 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 265 |
+
"loss": 0.0899,
|
| 266 |
+
"step": 43
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.34,
|
| 270 |
+
"learning_rate": 1.812987012987013e-05,
|
| 271 |
+
"loss": 0.1517,
|
| 272 |
+
"step": 44
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.34,
|
| 276 |
+
"learning_rate": 1.807792207792208e-05,
|
| 277 |
+
"loss": 0.1423,
|
| 278 |
+
"step": 45
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.35,
|
| 282 |
+
"learning_rate": 1.8025974025974027e-05,
|
| 283 |
+
"loss": 0.1418,
|
| 284 |
+
"step": 46
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.36,
|
| 288 |
+
"learning_rate": 1.7974025974025974e-05,
|
| 289 |
+
"loss": 0.1723,
|
| 290 |
+
"step": 47
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.37,
|
| 294 |
+
"learning_rate": 1.792207792207792e-05,
|
| 295 |
+
"loss": 0.1797,
|
| 296 |
+
"step": 48
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.37,
|
| 300 |
+
"learning_rate": 1.7870129870129872e-05,
|
| 301 |
+
"loss": 0.1365,
|
| 302 |
+
"step": 49
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.38,
|
| 306 |
+
"learning_rate": 1.781818181818182e-05,
|
| 307 |
+
"loss": 0.0738,
|
| 308 |
+
"step": 50
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.39,
|
| 312 |
+
"learning_rate": 1.7766233766233766e-05,
|
| 313 |
+
"loss": 0.0668,
|
| 314 |
+
"step": 51
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.4,
|
| 318 |
+
"learning_rate": 1.7714285714285717e-05,
|
| 319 |
+
"loss": 0.1373,
|
| 320 |
+
"step": 52
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.4,
|
| 324 |
+
"learning_rate": 1.7662337662337664e-05,
|
| 325 |
+
"loss": 0.0799,
|
| 326 |
+
"step": 53
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.41,
|
| 330 |
+
"learning_rate": 1.761038961038961e-05,
|
| 331 |
+
"loss": 0.1018,
|
| 332 |
+
"step": 54
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.42,
|
| 336 |
+
"learning_rate": 1.7558441558441558e-05,
|
| 337 |
+
"loss": 0.0802,
|
| 338 |
+
"step": 55
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.43,
|
| 342 |
+
"learning_rate": 1.750649350649351e-05,
|
| 343 |
+
"loss": 0.1481,
|
| 344 |
+
"step": 56
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.44,
|
| 348 |
+
"learning_rate": 1.7454545454545456e-05,
|
| 349 |
+
"loss": 0.0631,
|
| 350 |
+
"step": 57
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.44,
|
| 354 |
+
"learning_rate": 1.7402597402597403e-05,
|
| 355 |
+
"loss": 0.1118,
|
| 356 |
+
"step": 58
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.45,
|
| 360 |
+
"learning_rate": 1.735064935064935e-05,
|
| 361 |
+
"loss": 0.1086,
|
| 362 |
+
"step": 59
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.46,
|
| 366 |
+
"learning_rate": 1.72987012987013e-05,
|
| 367 |
+
"loss": 0.1337,
|
| 368 |
+
"step": 60
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.47,
|
| 372 |
+
"learning_rate": 1.7246753246753248e-05,
|
| 373 |
+
"loss": 0.1397,
|
| 374 |
+
"step": 61
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.47,
|
| 378 |
+
"learning_rate": 1.7194805194805195e-05,
|
| 379 |
+
"loss": 0.0519,
|
| 380 |
+
"step": 62
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.48,
|
| 384 |
+
"learning_rate": 1.7142857142857142e-05,
|
| 385 |
+
"loss": 0.1578,
|
| 386 |
+
"step": 63
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.49,
|
| 390 |
+
"learning_rate": 1.7090909090909092e-05,
|
| 391 |
+
"loss": 0.079,
|
| 392 |
+
"step": 64
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.5,
|
| 396 |
+
"learning_rate": 1.703896103896104e-05,
|
| 397 |
+
"loss": 0.1342,
|
| 398 |
+
"step": 65
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.5,
|
| 402 |
+
"learning_rate": 1.6987012987012987e-05,
|
| 403 |
+
"loss": 0.1528,
|
| 404 |
+
"step": 66
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.51,
|
| 408 |
+
"learning_rate": 1.6935064935064934e-05,
|
| 409 |
+
"loss": 0.1065,
|
| 410 |
+
"step": 67
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.52,
|
| 414 |
+
"learning_rate": 1.6883116883116884e-05,
|
| 415 |
+
"loss": 0.1988,
|
| 416 |
+
"step": 68
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.53,
|
| 420 |
+
"learning_rate": 1.683116883116883e-05,
|
| 421 |
+
"loss": 0.2357,
|
| 422 |
+
"step": 69
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.53,
|
| 426 |
+
"learning_rate": 1.677922077922078e-05,
|
| 427 |
+
"loss": 0.1789,
|
| 428 |
+
"step": 70
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.54,
|
| 432 |
+
"learning_rate": 1.672727272727273e-05,
|
| 433 |
+
"loss": 0.0914,
|
| 434 |
+
"step": 71
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.55,
|
| 438 |
+
"learning_rate": 1.6675324675324676e-05,
|
| 439 |
+
"loss": 0.1567,
|
| 440 |
+
"step": 72
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.56,
|
| 444 |
+
"learning_rate": 1.6623376623376627e-05,
|
| 445 |
+
"loss": 0.124,
|
| 446 |
+
"step": 73
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.56,
|
| 450 |
+
"learning_rate": 1.6571428571428574e-05,
|
| 451 |
+
"loss": 0.097,
|
| 452 |
+
"step": 74
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.57,
|
| 456 |
+
"learning_rate": 1.651948051948052e-05,
|
| 457 |
+
"loss": 0.087,
|
| 458 |
+
"step": 75
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.58,
|
| 462 |
+
"learning_rate": 1.646753246753247e-05,
|
| 463 |
+
"loss": 0.1105,
|
| 464 |
+
"step": 76
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.59,
|
| 468 |
+
"learning_rate": 1.641558441558442e-05,
|
| 469 |
+
"loss": 0.0719,
|
| 470 |
+
"step": 77
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.6,
|
| 474 |
+
"learning_rate": 1.6363636363636366e-05,
|
| 475 |
+
"loss": 0.0637,
|
| 476 |
+
"step": 78
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.6,
|
| 480 |
+
"learning_rate": 1.6311688311688313e-05,
|
| 481 |
+
"loss": 0.0794,
|
| 482 |
+
"step": 79
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.61,
|
| 486 |
+
"learning_rate": 1.6259740259740264e-05,
|
| 487 |
+
"loss": 0.125,
|
| 488 |
+
"step": 80
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.62,
|
| 492 |
+
"learning_rate": 1.620779220779221e-05,
|
| 493 |
+
"loss": 0.1411,
|
| 494 |
+
"step": 81
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.63,
|
| 498 |
+
"learning_rate": 1.6155844155844158e-05,
|
| 499 |
+
"loss": 0.0768,
|
| 500 |
+
"step": 82
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.63,
|
| 504 |
+
"learning_rate": 1.6103896103896105e-05,
|
| 505 |
+
"loss": 0.0986,
|
| 506 |
+
"step": 83
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.64,
|
| 510 |
+
"learning_rate": 1.6051948051948056e-05,
|
| 511 |
+
"loss": 0.0743,
|
| 512 |
+
"step": 84
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.65,
|
| 516 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 517 |
+
"loss": 0.0504,
|
| 518 |
+
"step": 85
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.66,
|
| 522 |
+
"learning_rate": 1.594805194805195e-05,
|
| 523 |
+
"loss": 0.0588,
|
| 524 |
+
"step": 86
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.66,
|
| 528 |
+
"learning_rate": 1.5896103896103897e-05,
|
| 529 |
+
"loss": 0.0699,
|
| 530 |
+
"step": 87
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.67,
|
| 534 |
+
"learning_rate": 1.5844155844155847e-05,
|
| 535 |
+
"loss": 0.1118,
|
| 536 |
+
"step": 88
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.68,
|
| 540 |
+
"learning_rate": 1.5792207792207795e-05,
|
| 541 |
+
"loss": 0.0397,
|
| 542 |
+
"step": 89
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.69,
|
| 546 |
+
"learning_rate": 1.5740259740259742e-05,
|
| 547 |
+
"loss": 0.0709,
|
| 548 |
+
"step": 90
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.69,
|
| 552 |
+
"learning_rate": 1.568831168831169e-05,
|
| 553 |
+
"loss": 0.0757,
|
| 554 |
+
"step": 91
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.7,
|
| 558 |
+
"learning_rate": 1.563636363636364e-05,
|
| 559 |
+
"loss": 0.0669,
|
| 560 |
+
"step": 92
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.71,
|
| 564 |
+
"learning_rate": 1.5584415584415587e-05,
|
| 565 |
+
"loss": 0.1155,
|
| 566 |
+
"step": 93
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.72,
|
| 570 |
+
"learning_rate": 1.5532467532467534e-05,
|
| 571 |
+
"loss": 0.0716,
|
| 572 |
+
"step": 94
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.73,
|
| 576 |
+
"learning_rate": 1.548051948051948e-05,
|
| 577 |
+
"loss": 0.0836,
|
| 578 |
+
"step": 95
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.73,
|
| 582 |
+
"learning_rate": 1.542857142857143e-05,
|
| 583 |
+
"loss": 0.0805,
|
| 584 |
+
"step": 96
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.74,
|
| 588 |
+
"learning_rate": 1.537662337662338e-05,
|
| 589 |
+
"loss": 0.0519,
|
| 590 |
+
"step": 97
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.75,
|
| 594 |
+
"learning_rate": 1.5324675324675326e-05,
|
| 595 |
+
"loss": 0.0903,
|
| 596 |
+
"step": 98
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.76,
|
| 600 |
+
"learning_rate": 1.5272727272727276e-05,
|
| 601 |
+
"loss": 0.1128,
|
| 602 |
+
"step": 99
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.76,
|
| 606 |
+
"learning_rate": 1.5220779220779223e-05,
|
| 607 |
+
"loss": 0.0448,
|
| 608 |
+
"step": 100
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.77,
|
| 612 |
+
"learning_rate": 1.516883116883117e-05,
|
| 613 |
+
"loss": 0.0941,
|
| 614 |
+
"step": 101
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.78,
|
| 618 |
+
"learning_rate": 1.511688311688312e-05,
|
| 619 |
+
"loss": 0.0782,
|
| 620 |
+
"step": 102
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.79,
|
| 624 |
+
"learning_rate": 1.5064935064935066e-05,
|
| 625 |
+
"loss": 0.1422,
|
| 626 |
+
"step": 103
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.79,
|
| 630 |
+
"learning_rate": 1.5012987012987015e-05,
|
| 631 |
+
"loss": 0.1379,
|
| 632 |
+
"step": 104
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.8,
|
| 636 |
+
"learning_rate": 1.4961038961038962e-05,
|
| 637 |
+
"loss": 0.1574,
|
| 638 |
+
"step": 105
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.81,
|
| 642 |
+
"learning_rate": 1.4909090909090911e-05,
|
| 643 |
+
"loss": 0.0878,
|
| 644 |
+
"step": 106
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.82,
|
| 648 |
+
"learning_rate": 1.4857142857142858e-05,
|
| 649 |
+
"loss": 0.0585,
|
| 650 |
+
"step": 107
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.82,
|
| 654 |
+
"learning_rate": 1.4805194805194807e-05,
|
| 655 |
+
"loss": 0.0342,
|
| 656 |
+
"step": 108
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.83,
|
| 660 |
+
"learning_rate": 1.4753246753246754e-05,
|
| 661 |
+
"loss": 0.1312,
|
| 662 |
+
"step": 109
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.84,
|
| 666 |
+
"learning_rate": 1.4701298701298703e-05,
|
| 667 |
+
"loss": 0.0919,
|
| 668 |
+
"step": 110
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.85,
|
| 672 |
+
"learning_rate": 1.464935064935065e-05,
|
| 673 |
+
"loss": 0.0693,
|
| 674 |
+
"step": 111
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.85,
|
| 678 |
+
"learning_rate": 1.45974025974026e-05,
|
| 679 |
+
"loss": 0.0661,
|
| 680 |
+
"step": 112
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.86,
|
| 684 |
+
"learning_rate": 1.4545454545454546e-05,
|
| 685 |
+
"loss": 0.0699,
|
| 686 |
+
"step": 113
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.87,
|
| 690 |
+
"learning_rate": 1.4493506493506495e-05,
|
| 691 |
+
"loss": 0.0784,
|
| 692 |
+
"step": 114
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.88,
|
| 696 |
+
"learning_rate": 1.4441558441558442e-05,
|
| 697 |
+
"loss": 0.0854,
|
| 698 |
+
"step": 115
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.89,
|
| 702 |
+
"learning_rate": 1.4389610389610391e-05,
|
| 703 |
+
"loss": 0.1409,
|
| 704 |
+
"step": 116
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.89,
|
| 708 |
+
"learning_rate": 1.433766233766234e-05,
|
| 709 |
+
"loss": 0.0833,
|
| 710 |
+
"step": 117
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9,
|
| 714 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 715 |
+
"loss": 0.068,
|
| 716 |
+
"step": 118
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.91,
|
| 720 |
+
"learning_rate": 1.4233766233766236e-05,
|
| 721 |
+
"loss": 0.0484,
|
| 722 |
+
"step": 119
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.92,
|
| 726 |
+
"learning_rate": 1.4181818181818183e-05,
|
| 727 |
+
"loss": 0.1664,
|
| 728 |
+
"step": 120
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.92,
|
| 732 |
+
"learning_rate": 1.4129870129870132e-05,
|
| 733 |
+
"loss": 0.1496,
|
| 734 |
+
"step": 121
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.93,
|
| 738 |
+
"learning_rate": 1.4077922077922079e-05,
|
| 739 |
+
"loss": 0.0956,
|
| 740 |
+
"step": 122
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.94,
|
| 744 |
+
"learning_rate": 1.4025974025974028e-05,
|
| 745 |
+
"loss": 0.0733,
|
| 746 |
+
"step": 123
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.95,
|
| 750 |
+
"learning_rate": 1.3974025974025975e-05,
|
| 751 |
+
"loss": 0.0785,
|
| 752 |
+
"step": 124
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.95,
|
| 756 |
+
"learning_rate": 1.3922077922077924e-05,
|
| 757 |
+
"loss": 0.1201,
|
| 758 |
+
"step": 125
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.96,
|
| 762 |
+
"learning_rate": 1.3870129870129871e-05,
|
| 763 |
+
"loss": 0.0393,
|
| 764 |
+
"step": 126
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.97,
|
| 768 |
+
"learning_rate": 1.381818181818182e-05,
|
| 769 |
+
"loss": 0.1055,
|
| 770 |
+
"step": 127
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.98,
|
| 774 |
+
"learning_rate": 1.3766233766233767e-05,
|
| 775 |
+
"loss": 0.0835,
|
| 776 |
+
"step": 128
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.98,
|
| 780 |
+
"learning_rate": 1.3714285714285716e-05,
|
| 781 |
+
"loss": 0.043,
|
| 782 |
+
"step": 129
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.99,
|
| 786 |
+
"learning_rate": 1.3662337662337663e-05,
|
| 787 |
+
"loss": 0.2135,
|
| 788 |
+
"step": 130
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.0,
|
| 792 |
+
"learning_rate": 1.3610389610389612e-05,
|
| 793 |
+
"loss": 0.0353,
|
| 794 |
+
"step": 131
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.0,
|
| 798 |
+
"eval_loss": 0.0589701272547245,
|
| 799 |
+
"eval_pearson": 0.8326156177710631,
|
| 800 |
+
"eval_runtime": 3.859,
|
| 801 |
+
"eval_samples_per_second": 60.119,
|
| 802 |
+
"eval_spearmanr": 0.8224932979448152,
|
| 803 |
+
"eval_steps_per_second": 3.887,
|
| 804 |
+
"step": 131
|
| 805 |
+
}
|
| 806 |
+
],
|
| 807 |
+
"max_steps": 393,
|
| 808 |
+
"num_train_epochs": 3,
|
| 809 |
+
"total_flos": 275792205551616.0,
|
| 810 |
+
"trial_name": null,
|
| 811 |
+
"trial_params": null
|
| 812 |
+
}
|
checkpoint-131/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f21f9c775e5279cf13583c754eb26ad37aa615454d716a4d935aebd04a43056
|
| 3 |
+
size 3311
|
checkpoint-131/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "S2312dal/M4_MLM",
|
| 3 |
+
"activation": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"DistilBertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"hidden_dim": 3072,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0"
|
| 13 |
+
},
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"label2id": {
|
| 16 |
+
"LABEL_0": 0
|
| 17 |
+
},
|
| 18 |
+
"max_position_embeddings": 512,
|
| 19 |
+
"model_type": "distilbert",
|
| 20 |
+
"n_heads": 12,
|
| 21 |
+
"n_layers": 6,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"problem_type": "regression",
|
| 24 |
+
"qa_dropout": 0.1,
|
| 25 |
+
"seq_classif_dropout": 0.2,
|
| 26 |
+
"sinusoidal_pos_embds": false,
|
| 27 |
+
"tie_weights_": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.20.0",
|
| 30 |
+
"vocab_size": 30522
|
| 31 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e24a89fd74ece8f3fa590ef12e846792801810dbb15c7ac06b70db892a0250e5
|
| 3 |
+
size 267851249
|
runs/Jun18_08-13-22_03eed8418a32/1655540012.1337984/events.out.tfevents.1655540012.03eed8418a32.67.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df08eace4f5dd22801fbcd731d8cfb2182693d3684b32c78291b804e1b1c6e8c
|
| 3 |
+
size 5314
|
runs/Jun18_08-13-22_03eed8418a32/events.out.tfevents.1655540012.03eed8418a32.67.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e70a42567b61a638ec419da006884207f95dcb9422881cdafcc96e30bf87352
|
| 3 |
+
size 24137
|
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 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"do_lower_case": true,
|
| 4 |
+
"mask_token": "[MASK]",
|
| 5 |
+
"model_max_length": 512,
|
| 6 |
+
"name_or_path": "S2312dal/M4_MLM",
|
| 7 |
+
"pad_token": "[PAD]",
|
| 8 |
+
"sep_token": "[SEP]",
|
| 9 |
+
"special_tokens_map_file": null,
|
| 10 |
+
"strip_accents": null,
|
| 11 |
+
"tokenize_chinese_chars": true,
|
| 12 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 13 |
+
"unk_token": "[UNK]"
|
| 14 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f21f9c775e5279cf13583c754eb26ad37aa615454d716a4d935aebd04a43056
|
| 3 |
+
size 3311
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|