Commit
•
2de16a2
1
Parent(s):
437010f
Upload folder using huggingface_hub
Browse files- README.md +36 -0
- checkpoint-4868/config.json +49 -0
- checkpoint-4868/model.safetensors +3 -0
- checkpoint-4868/optimizer.pt +3 -0
- checkpoint-4868/rng_state_0.pth +3 -0
- checkpoint-4868/rng_state_1.pth +3 -0
- checkpoint-4868/scheduler.pt +3 -0
- checkpoint-4868/trainer_state.json +1436 -0
- checkpoint-4868/training_args.bin +3 -0
- config.json +49 -0
- model.safetensors +3 -0
- runs/Nov09_20-41-49_6604152ce143/events.out.tfevents.1731184911.6604152ce143.302.0 +2 -2
- runs/Nov09_20-41-49_6604152ce143/events.out.tfevents.1731192961.6604152ce143.302.1 +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +57 -0
- training_args.bin +3 -0
- training_params.json +30 -0
- vocab.txt +0 -0
README.md
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
tags:
|
| 4 |
+
- autotrain
|
| 5 |
+
- text-classification
|
| 6 |
+
base_model: sismetanin/sbert-ru-sentiment-rusentiment
|
| 7 |
+
widget:
|
| 8 |
+
- text: "I love AutoTrain"
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# Model Trained Using AutoTrain
|
| 12 |
+
|
| 13 |
+
- Problem type: Text Classification
|
| 14 |
+
|
| 15 |
+
## Validation Metrics
|
| 16 |
+
loss: 0.8407641053199768
|
| 17 |
+
|
| 18 |
+
f1_macro: 0.4599197233920555
|
| 19 |
+
|
| 20 |
+
f1_micro: 0.6719408198910921
|
| 21 |
+
|
| 22 |
+
f1_weighted: 0.647985493521427
|
| 23 |
+
|
| 24 |
+
precision_macro: 0.4574685780248835
|
| 25 |
+
|
| 26 |
+
precision_micro: 0.6719408198910921
|
| 27 |
+
|
| 28 |
+
precision_weighted: 0.629823305374442
|
| 29 |
+
|
| 30 |
+
recall_macro: 0.46712379058109665
|
| 31 |
+
|
| 32 |
+
recall_micro: 0.6719408198910921
|
| 33 |
+
|
| 34 |
+
recall_weighted: 0.6719408198910921
|
| 35 |
+
|
| 36 |
+
accuracy: 0.6719408198910921
|
checkpoint-4868/config.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "sismetanin/sbert-ru-sentiment-rusentiment",
|
| 3 |
+
"_num_labels": 5,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"BertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.1,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"directionality": "bidi",
|
| 10 |
+
"finetuning_task": "sst2",
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.1,
|
| 14 |
+
"hidden_size": 1024,
|
| 15 |
+
"id2label": {
|
| 16 |
+
"0": "1",
|
| 17 |
+
"1": "2",
|
| 18 |
+
"2": "3",
|
| 19 |
+
"3": "4",
|
| 20 |
+
"4": "5"
|
| 21 |
+
},
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 4096,
|
| 24 |
+
"label2id": {
|
| 25 |
+
"1": 0,
|
| 26 |
+
"2": 1,
|
| 27 |
+
"3": 2,
|
| 28 |
+
"4": 3,
|
| 29 |
+
"5": 4
|
| 30 |
+
},
|
| 31 |
+
"layer_norm_eps": 1e-12,
|
| 32 |
+
"max_position_embeddings": 512,
|
| 33 |
+
"model_type": "bert",
|
| 34 |
+
"num_attention_heads": 16,
|
| 35 |
+
"num_hidden_layers": 24,
|
| 36 |
+
"pad_token_id": 0,
|
| 37 |
+
"pooler_fc_size": 768,
|
| 38 |
+
"pooler_num_attention_heads": 12,
|
| 39 |
+
"pooler_num_fc_layers": 3,
|
| 40 |
+
"pooler_size_per_head": 128,
|
| 41 |
+
"pooler_type": "first_token_transform",
|
| 42 |
+
"position_embedding_type": "absolute",
|
| 43 |
+
"problem_type": "single_label_classification",
|
| 44 |
+
"torch_dtype": "float32",
|
| 45 |
+
"transformers_version": "4.46.2",
|
| 46 |
+
"type_vocab_size": 2,
|
| 47 |
+
"use_cache": true,
|
| 48 |
+
"vocab_size": 120138
|
| 49 |
+
}
|
checkpoint-4868/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:59c091937666e74967004c1c61191959b87b68e70f47e11a004a23779935bd84
|
| 3 |
+
size 1707702436
|
checkpoint-4868/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8232e9ebc18f774e689b752fca099facfd57e16b2f1bc6d5f8acc6247c245f8b
|
| 3 |
+
size 3415639277
|
checkpoint-4868/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0a7ad84b056eef6f38438106ffd4be14c0680ecc3982120eb85a0364882f223
|
| 3 |
+
size 14512
|
checkpoint-4868/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37191115d6f064081317cd782733c92ab8957be89b7b332543236812c42cb69d
|
| 3 |
+
size 14512
|
checkpoint-4868/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24b420f93142fe76127ab30f1a728d9edc59fc5017adb75d8b4da9d71e2bf4a5
|
| 3 |
+
size 1064
|
checkpoint-4868/trainer_state.json
ADDED
|
@@ -0,0 +1,1436 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.064771056175232,
|
| 3 |
+
"best_model_checkpoint": "sbert-ru-huawei-sentiment-fine-up/checkpoint-4868",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4868,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.010271158586688579,
|
| 13 |
+
"grad_norm": 31.50909423828125,
|
| 14 |
+
"learning_rate": 5.476951163852123e-07,
|
| 15 |
+
"loss": 5.7521,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.020542317173377157,
|
| 20 |
+
"grad_norm": 65.67906951904297,
|
| 21 |
+
"learning_rate": 1.1182108626198083e-06,
|
| 22 |
+
"loss": 5.3431,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.030813475760065736,
|
| 27 |
+
"grad_norm": 47.638511657714844,
|
| 28 |
+
"learning_rate": 1.6659059790050208e-06,
|
| 29 |
+
"loss": 3.9751,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.041084634346754315,
|
| 34 |
+
"grad_norm": 23.86779022216797,
|
| 35 |
+
"learning_rate": 2.2364217252396165e-06,
|
| 36 |
+
"loss": 1.9721,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05135579293344289,
|
| 41 |
+
"grad_norm": 17.054542541503906,
|
| 42 |
+
"learning_rate": 2.806937471474213e-06,
|
| 43 |
+
"loss": 1.3281,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06162695152013147,
|
| 48 |
+
"grad_norm": 10.087335586547852,
|
| 49 |
+
"learning_rate": 3.3774532177088093e-06,
|
| 50 |
+
"loss": 1.236,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07189811010682005,
|
| 55 |
+
"grad_norm": 22.066057205200195,
|
| 56 |
+
"learning_rate": 3.9479689639434044e-06,
|
| 57 |
+
"loss": 1.0576,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08216926869350863,
|
| 62 |
+
"grad_norm": 16.504131317138672,
|
| 63 |
+
"learning_rate": 4.518484710178001e-06,
|
| 64 |
+
"loss": 0.9854,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0924404272801972,
|
| 69 |
+
"grad_norm": 16.981975555419922,
|
| 70 |
+
"learning_rate": 5.089000456412597e-06,
|
| 71 |
+
"loss": 0.9098,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.10271158586688578,
|
| 76 |
+
"grad_norm": 16.270986557006836,
|
| 77 |
+
"learning_rate": 5.659516202647193e-06,
|
| 78 |
+
"loss": 0.9133,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.11298274445357437,
|
| 83 |
+
"grad_norm": 13.069902420043945,
|
| 84 |
+
"learning_rate": 6.230031948881789e-06,
|
| 85 |
+
"loss": 0.8994,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.12325390304026294,
|
| 90 |
+
"grad_norm": 12.201191902160645,
|
| 91 |
+
"learning_rate": 6.800547695116385e-06,
|
| 92 |
+
"loss": 0.8731,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.13352506162695152,
|
| 97 |
+
"grad_norm": 7.924577236175537,
|
| 98 |
+
"learning_rate": 7.371063441350981e-06,
|
| 99 |
+
"loss": 0.9483,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1437962202136401,
|
| 104 |
+
"grad_norm": 16.856870651245117,
|
| 105 |
+
"learning_rate": 7.941579187585578e-06,
|
| 106 |
+
"loss": 0.7615,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.15406737880032867,
|
| 111 |
+
"grad_norm": 17.550830841064453,
|
| 112 |
+
"learning_rate": 8.512094933820173e-06,
|
| 113 |
+
"loss": 0.8699,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.16433853738701726,
|
| 118 |
+
"grad_norm": 8.01524543762207,
|
| 119 |
+
"learning_rate": 9.08261068005477e-06,
|
| 120 |
+
"loss": 0.9946,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.17460969597370585,
|
| 125 |
+
"grad_norm": 12.508491516113281,
|
| 126 |
+
"learning_rate": 9.653126426289367e-06,
|
| 127 |
+
"loss": 0.8598,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.1848808545603944,
|
| 132 |
+
"grad_norm": 11.055355072021484,
|
| 133 |
+
"learning_rate": 1.0223642172523962e-05,
|
| 134 |
+
"loss": 0.8902,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.195152013147083,
|
| 139 |
+
"grad_norm": 12.723591804504395,
|
| 140 |
+
"learning_rate": 1.0794157918758559e-05,
|
| 141 |
+
"loss": 0.9183,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.20542317173377156,
|
| 146 |
+
"grad_norm": 29.50878143310547,
|
| 147 |
+
"learning_rate": 1.1364673664993154e-05,
|
| 148 |
+
"loss": 0.9221,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.21569433032046015,
|
| 153 |
+
"grad_norm": 15.626035690307617,
|
| 154 |
+
"learning_rate": 1.193518941122775e-05,
|
| 155 |
+
"loss": 0.8211,
|
| 156 |
+
"step": 525
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.22596548890714874,
|
| 160 |
+
"grad_norm": 10.57879638671875,
|
| 161 |
+
"learning_rate": 1.2505705157462347e-05,
|
| 162 |
+
"loss": 0.9007,
|
| 163 |
+
"step": 550
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2362366474938373,
|
| 167 |
+
"grad_norm": 10.496281623840332,
|
| 168 |
+
"learning_rate": 1.3076220903696942e-05,
|
| 169 |
+
"loss": 0.8799,
|
| 170 |
+
"step": 575
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.2465078060805259,
|
| 174 |
+
"grad_norm": 12.28962516784668,
|
| 175 |
+
"learning_rate": 1.3646736649931538e-05,
|
| 176 |
+
"loss": 0.8934,
|
| 177 |
+
"step": 600
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.25677896466721445,
|
| 181 |
+
"grad_norm": 10.393588066101074,
|
| 182 |
+
"learning_rate": 1.4217252396166134e-05,
|
| 183 |
+
"loss": 0.8276,
|
| 184 |
+
"step": 625
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.26705012325390304,
|
| 188 |
+
"grad_norm": 11.521328926086426,
|
| 189 |
+
"learning_rate": 1.4787768142400731e-05,
|
| 190 |
+
"loss": 0.8975,
|
| 191 |
+
"step": 650
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2773212818405916,
|
| 195 |
+
"grad_norm": 10.707319259643555,
|
| 196 |
+
"learning_rate": 1.5358283888635328e-05,
|
| 197 |
+
"loss": 0.8862,
|
| 198 |
+
"step": 675
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.2875924404272802,
|
| 202 |
+
"grad_norm": 11.027580261230469,
|
| 203 |
+
"learning_rate": 1.5928799634869923e-05,
|
| 204 |
+
"loss": 0.839,
|
| 205 |
+
"step": 700
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.29786359901396875,
|
| 209 |
+
"grad_norm": 8.072233200073242,
|
| 210 |
+
"learning_rate": 1.6499315381104518e-05,
|
| 211 |
+
"loss": 0.8836,
|
| 212 |
+
"step": 725
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.30813475760065734,
|
| 216 |
+
"grad_norm": 5.254186153411865,
|
| 217 |
+
"learning_rate": 1.7069831127339113e-05,
|
| 218 |
+
"loss": 0.8102,
|
| 219 |
+
"step": 750
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3184059161873459,
|
| 223 |
+
"grad_norm": 9.644079208374023,
|
| 224 |
+
"learning_rate": 1.7640346873573712e-05,
|
| 225 |
+
"loss": 0.8363,
|
| 226 |
+
"step": 775
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.3286770747740345,
|
| 230 |
+
"grad_norm": 17.169221878051758,
|
| 231 |
+
"learning_rate": 1.8210862619808307e-05,
|
| 232 |
+
"loss": 0.7916,
|
| 233 |
+
"step": 800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.3389482333607231,
|
| 237 |
+
"grad_norm": 12.09295654296875,
|
| 238 |
+
"learning_rate": 1.8781378366042902e-05,
|
| 239 |
+
"loss": 0.8846,
|
| 240 |
+
"step": 825
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3492193919474117,
|
| 244 |
+
"grad_norm": 20.629968643188477,
|
| 245 |
+
"learning_rate": 1.93518941122775e-05,
|
| 246 |
+
"loss": 0.9433,
|
| 247 |
+
"step": 850
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.35949055053410023,
|
| 251 |
+
"grad_norm": 12.02403736114502,
|
| 252 |
+
"learning_rate": 1.9922409858512096e-05,
|
| 253 |
+
"loss": 0.7863,
|
| 254 |
+
"step": 875
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.3697617091207888,
|
| 258 |
+
"grad_norm": 7.93563985824585,
|
| 259 |
+
"learning_rate": 2.0492925604746694e-05,
|
| 260 |
+
"loss": 0.8346,
|
| 261 |
+
"step": 900
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3800328677074774,
|
| 265 |
+
"grad_norm": 10.790715217590332,
|
| 266 |
+
"learning_rate": 2.106344135098129e-05,
|
| 267 |
+
"loss": 0.8584,
|
| 268 |
+
"step": 925
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.390304026294166,
|
| 272 |
+
"grad_norm": 13.152490615844727,
|
| 273 |
+
"learning_rate": 2.1633957097215885e-05,
|
| 274 |
+
"loss": 0.8276,
|
| 275 |
+
"step": 950
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.4005751848808546,
|
| 279 |
+
"grad_norm": 8.368447303771973,
|
| 280 |
+
"learning_rate": 2.220447284345048e-05,
|
| 281 |
+
"loss": 0.9162,
|
| 282 |
+
"step": 975
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.4108463434675431,
|
| 286 |
+
"grad_norm": 5.231266021728516,
|
| 287 |
+
"learning_rate": 2.2774988589685075e-05,
|
| 288 |
+
"loss": 0.8999,
|
| 289 |
+
"step": 1000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.4211175020542317,
|
| 293 |
+
"grad_norm": 4.503598690032959,
|
| 294 |
+
"learning_rate": 2.334550433591967e-05,
|
| 295 |
+
"loss": 0.8748,
|
| 296 |
+
"step": 1025
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.4313886606409203,
|
| 300 |
+
"grad_norm": 5.98704719543457,
|
| 301 |
+
"learning_rate": 2.391602008215427e-05,
|
| 302 |
+
"loss": 0.9061,
|
| 303 |
+
"step": 1050
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.4416598192276089,
|
| 307 |
+
"grad_norm": 9.622344970703125,
|
| 308 |
+
"learning_rate": 2.4486535828388864e-05,
|
| 309 |
+
"loss": 0.9336,
|
| 310 |
+
"step": 1075
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.4519309778142975,
|
| 314 |
+
"grad_norm": 7.320569038391113,
|
| 315 |
+
"learning_rate": 2.5057051574623462e-05,
|
| 316 |
+
"loss": 0.8711,
|
| 317 |
+
"step": 1100
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.462202136400986,
|
| 321 |
+
"grad_norm": 6.236270427703857,
|
| 322 |
+
"learning_rate": 2.562756732085806e-05,
|
| 323 |
+
"loss": 0.8359,
|
| 324 |
+
"step": 1125
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.4724732949876746,
|
| 328 |
+
"grad_norm": 9.492266654968262,
|
| 329 |
+
"learning_rate": 2.6198083067092656e-05,
|
| 330 |
+
"loss": 0.8248,
|
| 331 |
+
"step": 1150
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.4827444535743632,
|
| 335 |
+
"grad_norm": 9.270378112792969,
|
| 336 |
+
"learning_rate": 2.676859881332725e-05,
|
| 337 |
+
"loss": 0.8539,
|
| 338 |
+
"step": 1175
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.4930156121610518,
|
| 342 |
+
"grad_norm": 8.744667053222656,
|
| 343 |
+
"learning_rate": 2.7316293929712462e-05,
|
| 344 |
+
"loss": 0.9035,
|
| 345 |
+
"step": 1200
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.5032867707477403,
|
| 349 |
+
"grad_norm": 7.469476222991943,
|
| 350 |
+
"learning_rate": 2.7886809675947057e-05,
|
| 351 |
+
"loss": 0.8951,
|
| 352 |
+
"step": 1225
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5135579293344289,
|
| 356 |
+
"grad_norm": 6.202262878417969,
|
| 357 |
+
"learning_rate": 2.8457325422181653e-05,
|
| 358 |
+
"loss": 0.9175,
|
| 359 |
+
"step": 1250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.5238290879211175,
|
| 363 |
+
"grad_norm": 6.080565452575684,
|
| 364 |
+
"learning_rate": 2.9027841168416248e-05,
|
| 365 |
+
"loss": 0.8948,
|
| 366 |
+
"step": 1275
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.5341002465078061,
|
| 370 |
+
"grad_norm": 10.15191650390625,
|
| 371 |
+
"learning_rate": 2.9598356914650843e-05,
|
| 372 |
+
"loss": 0.8631,
|
| 373 |
+
"step": 1300
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5443714050944947,
|
| 377 |
+
"grad_norm": 13.219573020935059,
|
| 378 |
+
"learning_rate": 3.0168872660885445e-05,
|
| 379 |
+
"loss": 0.9074,
|
| 380 |
+
"step": 1325
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5546425636811833,
|
| 384 |
+
"grad_norm": 7.929615497589111,
|
| 385 |
+
"learning_rate": 3.073938840712004e-05,
|
| 386 |
+
"loss": 0.8134,
|
| 387 |
+
"step": 1350
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.5649137222678718,
|
| 391 |
+
"grad_norm": 9.197999954223633,
|
| 392 |
+
"learning_rate": 3.130990415335464e-05,
|
| 393 |
+
"loss": 0.8826,
|
| 394 |
+
"step": 1375
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.5751848808545604,
|
| 398 |
+
"grad_norm": 9.510801315307617,
|
| 399 |
+
"learning_rate": 3.1880419899589233e-05,
|
| 400 |
+
"loss": 0.865,
|
| 401 |
+
"step": 1400
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.585456039441249,
|
| 405 |
+
"grad_norm": 14.126985549926758,
|
| 406 |
+
"learning_rate": 3.245093564582383e-05,
|
| 407 |
+
"loss": 0.9076,
|
| 408 |
+
"step": 1425
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5957271980279375,
|
| 412 |
+
"grad_norm": 13.595422744750977,
|
| 413 |
+
"learning_rate": 3.3021451392058424e-05,
|
| 414 |
+
"loss": 0.8754,
|
| 415 |
+
"step": 1450
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.6059983566146261,
|
| 419 |
+
"grad_norm": 8.661046028137207,
|
| 420 |
+
"learning_rate": 3.359196713829302e-05,
|
| 421 |
+
"loss": 0.8723,
|
| 422 |
+
"step": 1475
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.6162695152013147,
|
| 426 |
+
"grad_norm": 13.609139442443848,
|
| 427 |
+
"learning_rate": 3.4162482884527614e-05,
|
| 428 |
+
"loss": 0.8326,
|
| 429 |
+
"step": 1500
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.6265406737880033,
|
| 433 |
+
"grad_norm": 14.61794662475586,
|
| 434 |
+
"learning_rate": 3.473299863076221e-05,
|
| 435 |
+
"loss": 0.8022,
|
| 436 |
+
"step": 1525
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.6368118323746919,
|
| 440 |
+
"grad_norm": 10.622493743896484,
|
| 441 |
+
"learning_rate": 3.5303514376996804e-05,
|
| 442 |
+
"loss": 0.8726,
|
| 443 |
+
"step": 1550
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6470829909613804,
|
| 447 |
+
"grad_norm": 5.068788051605225,
|
| 448 |
+
"learning_rate": 3.585120949338202e-05,
|
| 449 |
+
"loss": 0.8936,
|
| 450 |
+
"step": 1575
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.657354149548069,
|
| 454 |
+
"grad_norm": 6.1167402267456055,
|
| 455 |
+
"learning_rate": 3.6421725239616614e-05,
|
| 456 |
+
"loss": 0.8834,
|
| 457 |
+
"step": 1600
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.6676253081347576,
|
| 461 |
+
"grad_norm": 6.667758941650391,
|
| 462 |
+
"learning_rate": 3.699224098585121e-05,
|
| 463 |
+
"loss": 0.925,
|
| 464 |
+
"step": 1625
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.6778964667214462,
|
| 468 |
+
"grad_norm": 11.814676284790039,
|
| 469 |
+
"learning_rate": 3.7562756732085804e-05,
|
| 470 |
+
"loss": 0.8439,
|
| 471 |
+
"step": 1650
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6881676253081348,
|
| 475 |
+
"grad_norm": 13.39101505279541,
|
| 476 |
+
"learning_rate": 3.8133272478320406e-05,
|
| 477 |
+
"loss": 0.7441,
|
| 478 |
+
"step": 1675
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6984387838948234,
|
| 482 |
+
"grad_norm": 7.23984956741333,
|
| 483 |
+
"learning_rate": 3.8703788224555e-05,
|
| 484 |
+
"loss": 0.939,
|
| 485 |
+
"step": 1700
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.7087099424815119,
|
| 489 |
+
"grad_norm": 16.45111846923828,
|
| 490 |
+
"learning_rate": 3.9274303970789596e-05,
|
| 491 |
+
"loss": 0.9217,
|
| 492 |
+
"step": 1725
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7189811010682005,
|
| 496 |
+
"grad_norm": 13.015716552734375,
|
| 497 |
+
"learning_rate": 3.984481971702419e-05,
|
| 498 |
+
"loss": 0.9648,
|
| 499 |
+
"step": 1750
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.729252259654889,
|
| 503 |
+
"grad_norm": 4.811708927154541,
|
| 504 |
+
"learning_rate": 4.041533546325879e-05,
|
| 505 |
+
"loss": 0.854,
|
| 506 |
+
"step": 1775
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.7395234182415776,
|
| 510 |
+
"grad_norm": 12.734000205993652,
|
| 511 |
+
"learning_rate": 4.098585120949339e-05,
|
| 512 |
+
"loss": 0.8663,
|
| 513 |
+
"step": 1800
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.7497945768282662,
|
| 517 |
+
"grad_norm": 5.706826210021973,
|
| 518 |
+
"learning_rate": 4.1556366955727984e-05,
|
| 519 |
+
"loss": 0.9436,
|
| 520 |
+
"step": 1825
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.7600657354149548,
|
| 524 |
+
"grad_norm": 10.904698371887207,
|
| 525 |
+
"learning_rate": 4.212688270196258e-05,
|
| 526 |
+
"loss": 0.9147,
|
| 527 |
+
"step": 1850
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.7703368940016434,
|
| 531 |
+
"grad_norm": 13.604780197143555,
|
| 532 |
+
"learning_rate": 4.2697398448197174e-05,
|
| 533 |
+
"loss": 0.9512,
|
| 534 |
+
"step": 1875
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.780608052588332,
|
| 538 |
+
"grad_norm": 11.89741325378418,
|
| 539 |
+
"learning_rate": 4.326791419443177e-05,
|
| 540 |
+
"loss": 0.9356,
|
| 541 |
+
"step": 1900
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.7908792111750206,
|
| 545 |
+
"grad_norm": 9.004523277282715,
|
| 546 |
+
"learning_rate": 4.3838429940666364e-05,
|
| 547 |
+
"loss": 0.927,
|
| 548 |
+
"step": 1925
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.8011503697617092,
|
| 552 |
+
"grad_norm": 11.654400825500488,
|
| 553 |
+
"learning_rate": 4.440894568690096e-05,
|
| 554 |
+
"loss": 0.8424,
|
| 555 |
+
"step": 1950
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.8114215283483976,
|
| 559 |
+
"grad_norm": 8.863574028015137,
|
| 560 |
+
"learning_rate": 4.4979461433135554e-05,
|
| 561 |
+
"loss": 0.8867,
|
| 562 |
+
"step": 1975
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.8216926869350862,
|
| 566 |
+
"grad_norm": 6.209610462188721,
|
| 567 |
+
"learning_rate": 4.554997717937015e-05,
|
| 568 |
+
"loss": 0.8153,
|
| 569 |
+
"step": 2000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.8319638455217748,
|
| 573 |
+
"grad_norm": 7.504034996032715,
|
| 574 |
+
"learning_rate": 4.6120492925604745e-05,
|
| 575 |
+
"loss": 0.855,
|
| 576 |
+
"step": 2025
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.8422350041084634,
|
| 580 |
+
"grad_norm": 12.570643424987793,
|
| 581 |
+
"learning_rate": 4.669100867183934e-05,
|
| 582 |
+
"loss": 0.9577,
|
| 583 |
+
"step": 2050
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.852506162695152,
|
| 587 |
+
"grad_norm": 9.295639038085938,
|
| 588 |
+
"learning_rate": 4.726152441807394e-05,
|
| 589 |
+
"loss": 0.84,
|
| 590 |
+
"step": 2075
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.8627773212818406,
|
| 594 |
+
"grad_norm": 8.446712493896484,
|
| 595 |
+
"learning_rate": 4.783204016430854e-05,
|
| 596 |
+
"loss": 0.9405,
|
| 597 |
+
"step": 2100
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.8730484798685292,
|
| 601 |
+
"grad_norm": 9.992650032043457,
|
| 602 |
+
"learning_rate": 4.840255591054313e-05,
|
| 603 |
+
"loss": 1.0059,
|
| 604 |
+
"step": 2125
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.8833196384552178,
|
| 608 |
+
"grad_norm": 9.760039329528809,
|
| 609 |
+
"learning_rate": 4.897307165677773e-05,
|
| 610 |
+
"loss": 0.8583,
|
| 611 |
+
"step": 2150
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.8935907970419064,
|
| 615 |
+
"grad_norm": 17.948223114013672,
|
| 616 |
+
"learning_rate": 4.954358740301233e-05,
|
| 617 |
+
"loss": 0.9798,
|
| 618 |
+
"step": 2175
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.903861955628595,
|
| 622 |
+
"grad_norm": 10.00885009765625,
|
| 623 |
+
"learning_rate": 4.9999992064830994e-05,
|
| 624 |
+
"loss": 1.1331,
|
| 625 |
+
"step": 2200
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9141331142152835,
|
| 629 |
+
"grad_norm": 9.577177047729492,
|
| 630 |
+
"learning_rate": 4.9999714334444495e-05,
|
| 631 |
+
"loss": 1.1493,
|
| 632 |
+
"step": 2225
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.924404272801972,
|
| 636 |
+
"grad_norm": 12.528915405273438,
|
| 637 |
+
"learning_rate": 4.999903985064472e-05,
|
| 638 |
+
"loss": 1.1891,
|
| 639 |
+
"step": 2250
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9346754313886606,
|
| 643 |
+
"grad_norm": 21.42876625061035,
|
| 644 |
+
"learning_rate": 4.9997968624135974e-05,
|
| 645 |
+
"loss": 1.1592,
|
| 646 |
+
"step": 2275
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.9449465899753492,
|
| 650 |
+
"grad_norm": 4.62033748626709,
|
| 651 |
+
"learning_rate": 4.9996500671918936e-05,
|
| 652 |
+
"loss": 1.1924,
|
| 653 |
+
"step": 2300
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.9552177485620378,
|
| 657 |
+
"grad_norm": 12.126282691955566,
|
| 658 |
+
"learning_rate": 4.999463601729047e-05,
|
| 659 |
+
"loss": 1.1318,
|
| 660 |
+
"step": 2325
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.9654889071487264,
|
| 664 |
+
"grad_norm": 7.639413356781006,
|
| 665 |
+
"learning_rate": 4.999237468984326e-05,
|
| 666 |
+
"loss": 1.1486,
|
| 667 |
+
"step": 2350
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.975760065735415,
|
| 671 |
+
"grad_norm": 7.761475563049316,
|
| 672 |
+
"learning_rate": 4.998971672546527e-05,
|
| 673 |
+
"loss": 1.1463,
|
| 674 |
+
"step": 2375
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.9860312243221035,
|
| 678 |
+
"grad_norm": 9.19140911102295,
|
| 679 |
+
"learning_rate": 4.998666216633926e-05,
|
| 680 |
+
"loss": 1.0832,
|
| 681 |
+
"step": 2400
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.9963023829087921,
|
| 685 |
+
"grad_norm": 6.729145526885986,
|
| 686 |
+
"learning_rate": 4.998321106094204e-05,
|
| 687 |
+
"loss": 1.0749,
|
| 688 |
+
"step": 2425
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.0,
|
| 692 |
+
"eval_accuracy": 0.5863556971129148,
|
| 693 |
+
"eval_f1_macro": 0.29454395983476717,
|
| 694 |
+
"eval_f1_micro": 0.5863556971129148,
|
| 695 |
+
"eval_f1_weighted": 0.47959265391783995,
|
| 696 |
+
"eval_loss": 1.1192505359649658,
|
| 697 |
+
"eval_precision_macro": 0.4105745703977606,
|
| 698 |
+
"eval_precision_micro": 0.5863556971129148,
|
| 699 |
+
"eval_precision_weighted": 0.5196212057725517,
|
| 700 |
+
"eval_recall_macro": 0.3058262541410274,
|
| 701 |
+
"eval_recall_micro": 0.5863556971129148,
|
| 702 |
+
"eval_recall_weighted": 0.5863556971129148,
|
| 703 |
+
"eval_runtime": 58.2052,
|
| 704 |
+
"eval_samples_per_second": 167.219,
|
| 705 |
+
"eval_steps_per_second": 5.24,
|
| 706 |
+
"step": 2434
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 1.0065735414954806,
|
| 710 |
+
"grad_norm": 12.559377670288086,
|
| 711 |
+
"learning_rate": 4.997936346404375e-05,
|
| 712 |
+
"loss": 1.2554,
|
| 713 |
+
"step": 2450
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.0168447000821692,
|
| 717 |
+
"grad_norm": 5.791057586669922,
|
| 718 |
+
"learning_rate": 4.997511943670697e-05,
|
| 719 |
+
"loss": 1.084,
|
| 720 |
+
"step": 2475
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.0271158586688578,
|
| 724 |
+
"grad_norm": 13.0701322555542,
|
| 725 |
+
"learning_rate": 4.9970479046285785e-05,
|
| 726 |
+
"loss": 1.1482,
|
| 727 |
+
"step": 2500
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.0373870172555464,
|
| 731 |
+
"grad_norm": 51.21843338012695,
|
| 732 |
+
"learning_rate": 4.9965442366424646e-05,
|
| 733 |
+
"loss": 1.1803,
|
| 734 |
+
"step": 2525
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.047658175842235,
|
| 738 |
+
"grad_norm": 8.89924430847168,
|
| 739 |
+
"learning_rate": 4.996000947705727e-05,
|
| 740 |
+
"loss": 1.1225,
|
| 741 |
+
"step": 2550
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 1.0579293344289236,
|
| 745 |
+
"grad_norm": 9.04704761505127,
|
| 746 |
+
"learning_rate": 4.9954180464405345e-05,
|
| 747 |
+
"loss": 1.1617,
|
| 748 |
+
"step": 2575
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 1.0682004930156122,
|
| 752 |
+
"grad_norm": 7.195584297180176,
|
| 753 |
+
"learning_rate": 4.9947955420977144e-05,
|
| 754 |
+
"loss": 1.0882,
|
| 755 |
+
"step": 2600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.0784716516023007,
|
| 759 |
+
"grad_norm": 16.638315200805664,
|
| 760 |
+
"learning_rate": 4.994133444556609e-05,
|
| 761 |
+
"loss": 1.1194,
|
| 762 |
+
"step": 2625
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.0887428101889893,
|
| 766 |
+
"grad_norm": 4.319007396697998,
|
| 767 |
+
"learning_rate": 4.993431764324916e-05,
|
| 768 |
+
"loss": 1.1418,
|
| 769 |
+
"step": 2650
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.099013968775678,
|
| 773 |
+
"grad_norm": 17.74847984313965,
|
| 774 |
+
"learning_rate": 4.992690512538525e-05,
|
| 775 |
+
"loss": 1.0342,
|
| 776 |
+
"step": 2675
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.1092851273623665,
|
| 780 |
+
"grad_norm": 10.52570915222168,
|
| 781 |
+
"learning_rate": 4.991909700961337e-05,
|
| 782 |
+
"loss": 1.1285,
|
| 783 |
+
"step": 2700
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.119556285949055,
|
| 787 |
+
"grad_norm": 5.708019733428955,
|
| 788 |
+
"learning_rate": 4.991089341985079e-05,
|
| 789 |
+
"loss": 1.0229,
|
| 790 |
+
"step": 2725
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.1298274445357437,
|
| 794 |
+
"grad_norm": 21.305837631225586,
|
| 795 |
+
"learning_rate": 4.9902294486291104e-05,
|
| 796 |
+
"loss": 1.0598,
|
| 797 |
+
"step": 2750
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.1400986031224323,
|
| 801 |
+
"grad_norm": 23.91814422607422,
|
| 802 |
+
"learning_rate": 4.9893300345402105e-05,
|
| 803 |
+
"loss": 1.1705,
|
| 804 |
+
"step": 2775
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.1503697617091209,
|
| 808 |
+
"grad_norm": 8.894378662109375,
|
| 809 |
+
"learning_rate": 4.988391113992368e-05,
|
| 810 |
+
"loss": 1.0257,
|
| 811 |
+
"step": 2800
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.1606409202958095,
|
| 815 |
+
"grad_norm": 12.799375534057617,
|
| 816 |
+
"learning_rate": 4.987412701886549e-05,
|
| 817 |
+
"loss": 1.1463,
|
| 818 |
+
"step": 2825
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.170912078882498,
|
| 822 |
+
"grad_norm": 12.658759117126465,
|
| 823 |
+
"learning_rate": 4.9863948137504666e-05,
|
| 824 |
+
"loss": 1.1462,
|
| 825 |
+
"step": 2850
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 1.1811832374691864,
|
| 829 |
+
"grad_norm": 6.2793707847595215,
|
| 830 |
+
"learning_rate": 4.9853374657383275e-05,
|
| 831 |
+
"loss": 1.0928,
|
| 832 |
+
"step": 2875
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 1.191454396055875,
|
| 836 |
+
"grad_norm": 10.470609664916992,
|
| 837 |
+
"learning_rate": 4.9842406746305814e-05,
|
| 838 |
+
"loss": 1.0454,
|
| 839 |
+
"step": 2900
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 1.2017255546425636,
|
| 843 |
+
"grad_norm": 46.87442398071289,
|
| 844 |
+
"learning_rate": 4.9831044578336524e-05,
|
| 845 |
+
"loss": 1.1075,
|
| 846 |
+
"step": 2925
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 1.2119967132292522,
|
| 850 |
+
"grad_norm": 9.962231636047363,
|
| 851 |
+
"learning_rate": 4.981928833379661e-05,
|
| 852 |
+
"loss": 1.1,
|
| 853 |
+
"step": 2950
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 1.2222678718159408,
|
| 857 |
+
"grad_norm": 4.477356433868408,
|
| 858 |
+
"learning_rate": 4.9807138199261414e-05,
|
| 859 |
+
"loss": 1.0811,
|
| 860 |
+
"step": 2975
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 1.2325390304026294,
|
| 864 |
+
"grad_norm": 11.651799201965332,
|
| 865 |
+
"learning_rate": 4.9794594367557435e-05,
|
| 866 |
+
"loss": 1.024,
|
| 867 |
+
"step": 3000
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 1.242810188989318,
|
| 871 |
+
"grad_norm": 7.755206108093262,
|
| 872 |
+
"learning_rate": 4.978165703775927e-05,
|
| 873 |
+
"loss": 1.067,
|
| 874 |
+
"step": 3025
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 1.2530813475760065,
|
| 878 |
+
"grad_norm": 15.455122947692871,
|
| 879 |
+
"learning_rate": 4.9768326415186454e-05,
|
| 880 |
+
"loss": 1.0311,
|
| 881 |
+
"step": 3050
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 1.2633525061626951,
|
| 885 |
+
"grad_norm": 9.131962776184082,
|
| 886 |
+
"learning_rate": 4.975460271140021e-05,
|
| 887 |
+
"loss": 1.0401,
|
| 888 |
+
"step": 3075
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 1.2736236647493837,
|
| 892 |
+
"grad_norm": 29.119171142578125,
|
| 893 |
+
"learning_rate": 4.974048614420007e-05,
|
| 894 |
+
"loss": 1.0969,
|
| 895 |
+
"step": 3100
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 1.2838948233360723,
|
| 899 |
+
"grad_norm": 6.163785457611084,
|
| 900 |
+
"learning_rate": 4.972597693762045e-05,
|
| 901 |
+
"loss": 1.1278,
|
| 902 |
+
"step": 3125
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.2941659819227609,
|
| 906 |
+
"grad_norm": 10.012632369995117,
|
| 907 |
+
"learning_rate": 4.971107532192707e-05,
|
| 908 |
+
"loss": 1.0993,
|
| 909 |
+
"step": 3150
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.3044371405094495,
|
| 913 |
+
"grad_norm": 6.5291666984558105,
|
| 914 |
+
"learning_rate": 4.96957815336133e-05,
|
| 915 |
+
"loss": 1.0934,
|
| 916 |
+
"step": 3175
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 1.314708299096138,
|
| 920 |
+
"grad_norm": 18.37449836730957,
|
| 921 |
+
"learning_rate": 4.9680095815396435e-05,
|
| 922 |
+
"loss": 1.104,
|
| 923 |
+
"step": 3200
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 1.3249794576828267,
|
| 927 |
+
"grad_norm": 8.442584991455078,
|
| 928 |
+
"learning_rate": 4.9664018416213796e-05,
|
| 929 |
+
"loss": 1.0918,
|
| 930 |
+
"step": 3225
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 1.3352506162695152,
|
| 934 |
+
"grad_norm": 5.685943603515625,
|
| 935 |
+
"learning_rate": 4.9647549591218835e-05,
|
| 936 |
+
"loss": 1.0535,
|
| 937 |
+
"step": 3250
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 1.3455217748562038,
|
| 941 |
+
"grad_norm": 5.7307448387146,
|
| 942 |
+
"learning_rate": 4.963068960177701e-05,
|
| 943 |
+
"loss": 1.1604,
|
| 944 |
+
"step": 3275
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 1.3557929334428924,
|
| 948 |
+
"grad_norm": 4.045636177062988,
|
| 949 |
+
"learning_rate": 4.961343871546175e-05,
|
| 950 |
+
"loss": 1.0484,
|
| 951 |
+
"step": 3300
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 1.366064092029581,
|
| 955 |
+
"grad_norm": 6.996286869049072,
|
| 956 |
+
"learning_rate": 4.959579720605009e-05,
|
| 957 |
+
"loss": 1.1055,
|
| 958 |
+
"step": 3325
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 1.3763352506162696,
|
| 962 |
+
"grad_norm": 10.471301078796387,
|
| 963 |
+
"learning_rate": 4.957776535351838e-05,
|
| 964 |
+
"loss": 1.119,
|
| 965 |
+
"step": 3350
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 1.3866064092029582,
|
| 969 |
+
"grad_norm": 6.319746017456055,
|
| 970 |
+
"learning_rate": 4.955934344403786e-05,
|
| 971 |
+
"loss": 1.127,
|
| 972 |
+
"step": 3375
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 1.3968775677896468,
|
| 976 |
+
"grad_norm": 6.542300224304199,
|
| 977 |
+
"learning_rate": 4.954053176997009e-05,
|
| 978 |
+
"loss": 0.9722,
|
| 979 |
+
"step": 3400
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 1.4071487263763354,
|
| 983 |
+
"grad_norm": 7.353511810302734,
|
| 984 |
+
"learning_rate": 4.952133062986232e-05,
|
| 985 |
+
"loss": 1.0545,
|
| 986 |
+
"step": 3425
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 1.417419884963024,
|
| 990 |
+
"grad_norm": 6.178640842437744,
|
| 991 |
+
"learning_rate": 4.950174032844274e-05,
|
| 992 |
+
"loss": 1.022,
|
| 993 |
+
"step": 3450
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 1.4276910435497125,
|
| 997 |
+
"grad_norm": 7.4956955909729,
|
| 998 |
+
"learning_rate": 4.948176117661567e-05,
|
| 999 |
+
"loss": 1.0411,
|
| 1000 |
+
"step": 3475
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.437962202136401,
|
| 1004 |
+
"grad_norm": 16.34415626525879,
|
| 1005 |
+
"learning_rate": 4.9461393491456595e-05,
|
| 1006 |
+
"loss": 1.0635,
|
| 1007 |
+
"step": 3500
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 1.4482333607230895,
|
| 1011 |
+
"grad_norm": 16.984901428222656,
|
| 1012 |
+
"learning_rate": 4.944063759620715e-05,
|
| 1013 |
+
"loss": 1.1581,
|
| 1014 |
+
"step": 3525
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 1.458504519309778,
|
| 1018 |
+
"grad_norm": 17.63494300842285,
|
| 1019 |
+
"learning_rate": 4.941949382027e-05,
|
| 1020 |
+
"loss": 1.0962,
|
| 1021 |
+
"step": 3550
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 1.4687756778964667,
|
| 1025 |
+
"grad_norm": 5.165820598602295,
|
| 1026 |
+
"learning_rate": 4.939796249920358e-05,
|
| 1027 |
+
"loss": 1.0976,
|
| 1028 |
+
"step": 3575
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.4790468364831553,
|
| 1032 |
+
"grad_norm": 6.0279436111450195,
|
| 1033 |
+
"learning_rate": 4.93760439747168e-05,
|
| 1034 |
+
"loss": 1.0734,
|
| 1035 |
+
"step": 3600
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 1.4893179950698439,
|
| 1039 |
+
"grad_norm": 4.7154717445373535,
|
| 1040 |
+
"learning_rate": 4.935373859466361e-05,
|
| 1041 |
+
"loss": 1.0507,
|
| 1042 |
+
"step": 3625
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 1.4995891536565324,
|
| 1046 |
+
"grad_norm": 11.337488174438477,
|
| 1047 |
+
"learning_rate": 4.93310467130375e-05,
|
| 1048 |
+
"loss": 1.1746,
|
| 1049 |
+
"step": 3650
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 1.509860312243221,
|
| 1053 |
+
"grad_norm": 18.193342208862305,
|
| 1054 |
+
"learning_rate": 4.930796868996582e-05,
|
| 1055 |
+
"loss": 1.0982,
|
| 1056 |
+
"step": 3675
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 1.5201314708299096,
|
| 1060 |
+
"grad_norm": 7.017096996307373,
|
| 1061 |
+
"learning_rate": 4.928450489170415e-05,
|
| 1062 |
+
"loss": 1.1556,
|
| 1063 |
+
"step": 3700
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 1.5304026294165982,
|
| 1067 |
+
"grad_norm": 2.5547895431518555,
|
| 1068 |
+
"learning_rate": 4.926065569063041e-05,
|
| 1069 |
+
"loss": 1.0685,
|
| 1070 |
+
"step": 3725
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 1.5406737880032868,
|
| 1074 |
+
"grad_norm": 8.476696014404297,
|
| 1075 |
+
"learning_rate": 4.9236421465239025e-05,
|
| 1076 |
+
"loss": 1.0737,
|
| 1077 |
+
"step": 3750
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 1.5509449465899754,
|
| 1081 |
+
"grad_norm": 8.107565879821777,
|
| 1082 |
+
"learning_rate": 4.921180260013483e-05,
|
| 1083 |
+
"loss": 1.086,
|
| 1084 |
+
"step": 3775
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 1.561216105176664,
|
| 1088 |
+
"grad_norm": 8.92119312286377,
|
| 1089 |
+
"learning_rate": 4.9186799486027057e-05,
|
| 1090 |
+
"loss": 1.1056,
|
| 1091 |
+
"step": 3800
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 1.5714872637633523,
|
| 1095 |
+
"grad_norm": 9.719097137451172,
|
| 1096 |
+
"learning_rate": 4.916141251972307e-05,
|
| 1097 |
+
"loss": 0.9651,
|
| 1098 |
+
"step": 3825
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.581758422350041,
|
| 1102 |
+
"grad_norm": 16.80173110961914,
|
| 1103 |
+
"learning_rate": 4.9135642104122093e-05,
|
| 1104 |
+
"loss": 0.9173,
|
| 1105 |
+
"step": 3850
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 1.5920295809367295,
|
| 1109 |
+
"grad_norm": 15.660393714904785,
|
| 1110 |
+
"learning_rate": 4.910948864820882e-05,
|
| 1111 |
+
"loss": 1.0374,
|
| 1112 |
+
"step": 3875
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.6023007395234181,
|
| 1116 |
+
"grad_norm": 17.720876693725586,
|
| 1117 |
+
"learning_rate": 4.90829525670469e-05,
|
| 1118 |
+
"loss": 0.9644,
|
| 1119 |
+
"step": 3900
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 1.6125718981101067,
|
| 1123 |
+
"grad_norm": 31.2305965423584,
|
| 1124 |
+
"learning_rate": 4.905603428177238e-05,
|
| 1125 |
+
"loss": 1.0033,
|
| 1126 |
+
"step": 3925
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 1.6228430566967953,
|
| 1130 |
+
"grad_norm": 7.728428840637207,
|
| 1131 |
+
"learning_rate": 4.9028734219587e-05,
|
| 1132 |
+
"loss": 1.0823,
|
| 1133 |
+
"step": 3950
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 1.6331142152834839,
|
| 1137 |
+
"grad_norm": 15.458768844604492,
|
| 1138 |
+
"learning_rate": 4.900105281375143e-05,
|
| 1139 |
+
"loss": 1.0676,
|
| 1140 |
+
"step": 3975
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.6433853738701725,
|
| 1144 |
+
"grad_norm": 18.69976806640625,
|
| 1145 |
+
"learning_rate": 4.8972990503578386e-05,
|
| 1146 |
+
"loss": 1.0445,
|
| 1147 |
+
"step": 4000
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 1.653656532456861,
|
| 1151 |
+
"grad_norm": 8.92398452758789,
|
| 1152 |
+
"learning_rate": 4.894454773442564e-05,
|
| 1153 |
+
"loss": 1.0722,
|
| 1154 |
+
"step": 4025
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 1.6639276910435497,
|
| 1158 |
+
"grad_norm": 12.593587875366211,
|
| 1159 |
+
"learning_rate": 4.891572495768898e-05,
|
| 1160 |
+
"loss": 1.0598,
|
| 1161 |
+
"step": 4050
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 1.6741988496302382,
|
| 1165 |
+
"grad_norm": 4.575959205627441,
|
| 1166 |
+
"learning_rate": 4.8886522630795045e-05,
|
| 1167 |
+
"loss": 1.0429,
|
| 1168 |
+
"step": 4075
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 1.6844700082169268,
|
| 1172 |
+
"grad_norm": 14.307016372680664,
|
| 1173 |
+
"learning_rate": 4.8856941217194035e-05,
|
| 1174 |
+
"loss": 0.9699,
|
| 1175 |
+
"step": 4100
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 1.6947411668036154,
|
| 1179 |
+
"grad_norm": 6.749273300170898,
|
| 1180 |
+
"learning_rate": 4.882698118635239e-05,
|
| 1181 |
+
"loss": 1.0919,
|
| 1182 |
+
"step": 4125
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 1.705012325390304,
|
| 1186 |
+
"grad_norm": 9.383212089538574,
|
| 1187 |
+
"learning_rate": 4.879664301374532e-05,
|
| 1188 |
+
"loss": 1.1693,
|
| 1189 |
+
"step": 4150
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 1.7152834839769926,
|
| 1193 |
+
"grad_norm": 4.1049909591674805,
|
| 1194 |
+
"learning_rate": 4.876592718084928e-05,
|
| 1195 |
+
"loss": 1.159,
|
| 1196 |
+
"step": 4175
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 1.7255546425636812,
|
| 1200 |
+
"grad_norm": 8.32151985168457,
|
| 1201 |
+
"learning_rate": 4.873483417513429e-05,
|
| 1202 |
+
"loss": 1.0969,
|
| 1203 |
+
"step": 4200
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 1.7358258011503698,
|
| 1207 |
+
"grad_norm": 6.129182815551758,
|
| 1208 |
+
"learning_rate": 4.8703364490056245e-05,
|
| 1209 |
+
"loss": 1.0632,
|
| 1210 |
+
"step": 4225
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 1.7460969597370584,
|
| 1214 |
+
"grad_norm": 5.181979179382324,
|
| 1215 |
+
"learning_rate": 4.8671518625049036e-05,
|
| 1216 |
+
"loss": 0.9523,
|
| 1217 |
+
"step": 4250
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 1.756368118323747,
|
| 1221 |
+
"grad_norm": 6.566084384918213,
|
| 1222 |
+
"learning_rate": 4.863929708551667e-05,
|
| 1223 |
+
"loss": 1.0336,
|
| 1224 |
+
"step": 4275
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 1.7666392769104355,
|
| 1228 |
+
"grad_norm": 14.47562313079834,
|
| 1229 |
+
"learning_rate": 4.8606700382825224e-05,
|
| 1230 |
+
"loss": 1.0493,
|
| 1231 |
+
"step": 4300
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 1.7769104354971241,
|
| 1235 |
+
"grad_norm": 3.870394706726074,
|
| 1236 |
+
"learning_rate": 4.8573729034294736e-05,
|
| 1237 |
+
"loss": 1.1012,
|
| 1238 |
+
"step": 4325
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 1.7871815940838127,
|
| 1242 |
+
"grad_norm": 21.563486099243164,
|
| 1243 |
+
"learning_rate": 4.854038356319098e-05,
|
| 1244 |
+
"loss": 1.0629,
|
| 1245 |
+
"step": 4350
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 1.7974527526705013,
|
| 1249 |
+
"grad_norm": 4.163707256317139,
|
| 1250 |
+
"learning_rate": 4.850666449871718e-05,
|
| 1251 |
+
"loss": 1.0252,
|
| 1252 |
+
"step": 4375
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 1.80772391125719,
|
| 1256 |
+
"grad_norm": 7.478860378265381,
|
| 1257 |
+
"learning_rate": 4.8472572376005614e-05,
|
| 1258 |
+
"loss": 1.1695,
|
| 1259 |
+
"step": 4400
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 1.8179950698438785,
|
| 1263 |
+
"grad_norm": 8.34506893157959,
|
| 1264 |
+
"learning_rate": 4.843810773610911e-05,
|
| 1265 |
+
"loss": 1.0471,
|
| 1266 |
+
"step": 4425
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 1.828266228430567,
|
| 1270 |
+
"grad_norm": 7.001756191253662,
|
| 1271 |
+
"learning_rate": 4.840327112599244e-05,
|
| 1272 |
+
"loss": 1.0156,
|
| 1273 |
+
"step": 4450
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 1.8385373870172557,
|
| 1277 |
+
"grad_norm": 4.876439094543457,
|
| 1278 |
+
"learning_rate": 4.836806309852371e-05,
|
| 1279 |
+
"loss": 1.022,
|
| 1280 |
+
"step": 4475
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 1.8488085456039443,
|
| 1284 |
+
"grad_norm": 5.670014381408691,
|
| 1285 |
+
"learning_rate": 4.833248421246548e-05,
|
| 1286 |
+
"loss": 1.1103,
|
| 1287 |
+
"step": 4500
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 1.8590797041906328,
|
| 1291 |
+
"grad_norm": 5.291884422302246,
|
| 1292 |
+
"learning_rate": 4.8296535032466e-05,
|
| 1293 |
+
"loss": 1.0795,
|
| 1294 |
+
"step": 4525
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 1.8693508627773214,
|
| 1298 |
+
"grad_norm": 7.388947486877441,
|
| 1299 |
+
"learning_rate": 4.826021612905017e-05,
|
| 1300 |
+
"loss": 1.1053,
|
| 1301 |
+
"step": 4550
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 1.87962202136401,
|
| 1305 |
+
"grad_norm": 5.867747783660889,
|
| 1306 |
+
"learning_rate": 4.822352807861054e-05,
|
| 1307 |
+
"loss": 1.0175,
|
| 1308 |
+
"step": 4575
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 1.8898931799506984,
|
| 1312 |
+
"grad_norm": 11.747688293457031,
|
| 1313 |
+
"learning_rate": 4.818647146339813e-05,
|
| 1314 |
+
"loss": 1.0419,
|
| 1315 |
+
"step": 4600
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 1.900164338537387,
|
| 1319 |
+
"grad_norm": 7.364708423614502,
|
| 1320 |
+
"learning_rate": 4.814904687151321e-05,
|
| 1321 |
+
"loss": 1.0909,
|
| 1322 |
+
"step": 4625
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 1.9104354971240756,
|
| 1326 |
+
"grad_norm": 10.691391944885254,
|
| 1327 |
+
"learning_rate": 4.811125489689594e-05,
|
| 1328 |
+
"loss": 1.1139,
|
| 1329 |
+
"step": 4650
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.9207066557107642,
|
| 1333 |
+
"grad_norm": 61.839820861816406,
|
| 1334 |
+
"learning_rate": 4.807309613931698e-05,
|
| 1335 |
+
"loss": 1.0478,
|
| 1336 |
+
"step": 4675
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.9309778142974527,
|
| 1340 |
+
"grad_norm": 3.956354856491089,
|
| 1341 |
+
"learning_rate": 4.803457120436792e-05,
|
| 1342 |
+
"loss": 1.1058,
|
| 1343 |
+
"step": 4700
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 1.9412489728841413,
|
| 1347 |
+
"grad_norm": 8.450248718261719,
|
| 1348 |
+
"learning_rate": 4.799568070345175e-05,
|
| 1349 |
+
"loss": 1.1035,
|
| 1350 |
+
"step": 4725
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 1.95152013147083,
|
| 1354 |
+
"grad_norm": 9.974756240844727,
|
| 1355 |
+
"learning_rate": 4.795642525377307e-05,
|
| 1356 |
+
"loss": 1.0522,
|
| 1357 |
+
"step": 4750
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 1.9617912900575185,
|
| 1361 |
+
"grad_norm": 23.98746109008789,
|
| 1362 |
+
"learning_rate": 4.791680547832835e-05,
|
| 1363 |
+
"loss": 1.0571,
|
| 1364 |
+
"step": 4775
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 1.972062448644207,
|
| 1368 |
+
"grad_norm": 10.666019439697266,
|
| 1369 |
+
"learning_rate": 4.7876822005896016e-05,
|
| 1370 |
+
"loss": 1.1004,
|
| 1371 |
+
"step": 4800
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 1.9823336072308957,
|
| 1375 |
+
"grad_norm": 8.37453842163086,
|
| 1376 |
+
"learning_rate": 4.7836475471026484e-05,
|
| 1377 |
+
"loss": 1.106,
|
| 1378 |
+
"step": 4825
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 1.9926047658175843,
|
| 1382 |
+
"grad_norm": 6.877459526062012,
|
| 1383 |
+
"learning_rate": 4.779576651403209e-05,
|
| 1384 |
+
"loss": 1.0404,
|
| 1385 |
+
"step": 4850
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 2.0,
|
| 1389 |
+
"eval_accuracy": 0.6005342648720846,
|
| 1390 |
+
"eval_f1_macro": 0.34074175503832355,
|
| 1391 |
+
"eval_f1_micro": 0.6005342648720846,
|
| 1392 |
+
"eval_f1_weighted": 0.5316262488920555,
|
| 1393 |
+
"eval_loss": 1.064771056175232,
|
| 1394 |
+
"eval_precision_macro": 0.42047577765386696,
|
| 1395 |
+
"eval_precision_micro": 0.6005342648720846,
|
| 1396 |
+
"eval_precision_weighted": 0.5362395289892545,
|
| 1397 |
+
"eval_recall_macro": 0.32992208872704937,
|
| 1398 |
+
"eval_recall_micro": 0.6005342648720846,
|
| 1399 |
+
"eval_recall_weighted": 0.6005342648720846,
|
| 1400 |
+
"eval_runtime": 58.1251,
|
| 1401 |
+
"eval_samples_per_second": 167.449,
|
| 1402 |
+
"eval_steps_per_second": 5.247,
|
| 1403 |
+
"step": 4868
|
| 1404 |
+
}
|
| 1405 |
+
],
|
| 1406 |
+
"logging_steps": 25,
|
| 1407 |
+
"max_steps": 21906,
|
| 1408 |
+
"num_input_tokens_seen": 0,
|
| 1409 |
+
"num_train_epochs": 9,
|
| 1410 |
+
"save_steps": 500,
|
| 1411 |
+
"stateful_callbacks": {
|
| 1412 |
+
"EarlyStoppingCallback": {
|
| 1413 |
+
"args": {
|
| 1414 |
+
"early_stopping_patience": 5,
|
| 1415 |
+
"early_stopping_threshold": 0.01
|
| 1416 |
+
},
|
| 1417 |
+
"attributes": {
|
| 1418 |
+
"early_stopping_patience_counter": 0
|
| 1419 |
+
}
|
| 1420 |
+
},
|
| 1421 |
+
"TrainerControl": {
|
| 1422 |
+
"args": {
|
| 1423 |
+
"should_epoch_stop": false,
|
| 1424 |
+
"should_evaluate": false,
|
| 1425 |
+
"should_log": false,
|
| 1426 |
+
"should_save": true,
|
| 1427 |
+
"should_training_stop": false
|
| 1428 |
+
},
|
| 1429 |
+
"attributes": {}
|
| 1430 |
+
}
|
| 1431 |
+
},
|
| 1432 |
+
"total_flos": 3.629350229586739e+16,
|
| 1433 |
+
"train_batch_size": 8,
|
| 1434 |
+
"trial_name": null,
|
| 1435 |
+
"trial_params": null
|
| 1436 |
+
}
|
checkpoint-4868/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae1e732367dbea8de8f8eae10cfb26b3b4030d51ce844fed59cb0541be0639ad
|
| 3 |
+
size 5304
|
config.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "sismetanin/sbert-ru-sentiment-rusentiment",
|
| 3 |
+
"_num_labels": 5,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"BertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.1,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"directionality": "bidi",
|
| 10 |
+
"finetuning_task": "sst2",
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.1,
|
| 14 |
+
"hidden_size": 1024,
|
| 15 |
+
"id2label": {
|
| 16 |
+
"0": "1",
|
| 17 |
+
"1": "2",
|
| 18 |
+
"2": "3",
|
| 19 |
+
"3": "4",
|
| 20 |
+
"4": "5"
|
| 21 |
+
},
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 4096,
|
| 24 |
+
"label2id": {
|
| 25 |
+
"1": 0,
|
| 26 |
+
"2": 1,
|
| 27 |
+
"3": 2,
|
| 28 |
+
"4": 3,
|
| 29 |
+
"5": 4
|
| 30 |
+
},
|
| 31 |
+
"layer_norm_eps": 1e-12,
|
| 32 |
+
"max_position_embeddings": 512,
|
| 33 |
+
"model_type": "bert",
|
| 34 |
+
"num_attention_heads": 16,
|
| 35 |
+
"num_hidden_layers": 24,
|
| 36 |
+
"pad_token_id": 0,
|
| 37 |
+
"pooler_fc_size": 768,
|
| 38 |
+
"pooler_num_attention_heads": 12,
|
| 39 |
+
"pooler_num_fc_layers": 3,
|
| 40 |
+
"pooler_size_per_head": 128,
|
| 41 |
+
"pooler_type": "first_token_transform",
|
| 42 |
+
"position_embedding_type": "absolute",
|
| 43 |
+
"problem_type": "single_label_classification",
|
| 44 |
+
"torch_dtype": "float32",
|
| 45 |
+
"transformers_version": "4.46.2",
|
| 46 |
+
"type_vocab_size": 2,
|
| 47 |
+
"use_cache": true,
|
| 48 |
+
"vocab_size": 120138
|
| 49 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:59c091937666e74967004c1c61191959b87b68e70f47e11a004a23779935bd84
|
| 3 |
+
size 1707702436
|
runs/Nov09_20-41-49_6604152ce143/events.out.tfevents.1731184911.6604152ce143.302.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f76b6cccfa2d20a035705c700b73c20993d4bdfe38e9cb5d180d26727c53d242
|
| 3 |
+
size 155527
|
runs/Nov09_20-41-49_6604152ce143/events.out.tfevents.1731192961.6604152ce143.302.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea5b52a8fe19f62d16427f8e611b19d729b57b0b6ab307a50947a0d3a9e6ddea
|
| 3 |
+
size 936
|
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 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 50 |
+
"never_split": null,
|
| 51 |
+
"pad_token": "[PAD]",
|
| 52 |
+
"sep_token": "[SEP]",
|
| 53 |
+
"strip_accents": null,
|
| 54 |
+
"tokenize_chinese_chars": true,
|
| 55 |
+
"tokenizer_class": "BertTokenizer",
|
| 56 |
+
"unk_token": "[UNK]"
|
| 57 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae1e732367dbea8de8f8eae10cfb26b3b4030d51ce844fed59cb0541be0639ad
|
| 3 |
+
size 5304
|
training_params.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"data_path": "sbert-ru-huawei-sentiment-fine-up/autotrain-data",
|
| 3 |
+
"model": "sismetanin/sbert-ru-sentiment-rusentiment",
|
| 4 |
+
"lr": 5e-05,
|
| 5 |
+
"epochs": 9,
|
| 6 |
+
"max_seq_length": 256,
|
| 7 |
+
"batch_size": 8,
|
| 8 |
+
"warmup_ratio": 0.1,
|
| 9 |
+
"gradient_accumulation": 1,
|
| 10 |
+
"optimizer": "adamw_torch",
|
| 11 |
+
"scheduler": "cosine_with_restarts",
|
| 12 |
+
"weight_decay": 0.0,
|
| 13 |
+
"max_grad_norm": 1.0,
|
| 14 |
+
"seed": 42,
|
| 15 |
+
"train_split": "train",
|
| 16 |
+
"valid_split": "validation",
|
| 17 |
+
"text_column": "autotrain_text",
|
| 18 |
+
"target_column": "autotrain_label",
|
| 19 |
+
"logging_steps": -1,
|
| 20 |
+
"project_name": "sbert-ru-huawei-sentiment-fine-up",
|
| 21 |
+
"auto_find_batch_size": false,
|
| 22 |
+
"mixed_precision": "fp16",
|
| 23 |
+
"save_total_limit": 1,
|
| 24 |
+
"push_to_hub": true,
|
| 25 |
+
"eval_strategy": "epoch",
|
| 26 |
+
"username": "ivanovsdesign",
|
| 27 |
+
"log": "tensorboard",
|
| 28 |
+
"early_stopping_patience": 5,
|
| 29 |
+
"early_stopping_threshold": 0.01
|
| 30 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|