Gabriel Louzada da Silva commited on
Upload folder using huggingface_hub
Browse files- checkpoint-1610/config.json +42 -0
- checkpoint-1610/model.safetensors +3 -0
- checkpoint-1610/optimizer.pt +3 -0
- checkpoint-1610/rng_state.pth +3 -0
- checkpoint-1610/scheduler.pt +3 -0
- checkpoint-1610/trainer_state.json +1194 -0
- checkpoint-1610/training_args.bin +3 -0
- checkpoint-2412/config.json +42 -0
- checkpoint-2412/model.safetensors +3 -0
- checkpoint-2412/optimizer.pt +3 -0
- checkpoint-2412/rng_state.pth +3 -0
- checkpoint-2412/scheduler.pt +3 -0
- checkpoint-2412/trainer_state.json +1766 -0
- checkpoint-2412/training_args.bin +3 -0
- checkpoint-805/config.json +42 -0
- checkpoint-805/model.safetensors +3 -0
- checkpoint-805/optimizer.pt +3 -0
- checkpoint-805/rng_state.pth +3 -0
- checkpoint-805/scheduler.pt +3 -0
- checkpoint-805/trainer_state.json +615 -0
- checkpoint-805/training_args.bin +3 -0
- config.json +42 -0
- logs/events.out.tfevents.1743420092.DESKTOP-JBF3D6L.2192.0 +3 -0
- logs/events.out.tfevents.1743420252.DESKTOP-JBF3D6L.6468.0 +3 -0
- logs/events.out.tfevents.1743420302.DESKTOP-JBF3D6L.11468.0 +3 -0
- logs/events.out.tfevents.1743420355.DESKTOP-JBF3D6L.6752.0 +3 -0
- logs/events.out.tfevents.1743420703.DESKTOP-JBF3D6L.4224.0 +3 -0
- logs/events.out.tfevents.1743423112.DESKTOP-JBF3D6L.9152.0 +3 -0
- logs/events.out.tfevents.1743427318.DESKTOP-JBF3D6L.9152.1 +3 -0
- model.safetensors +3 -0
- special_tokens_map.json +7 -0
- tokenizer_config.json +58 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
checkpoint-1610/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"directionality": "bidi",
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"LABEL_0": 0,
|
| 20 |
+
"LABEL_1": 1,
|
| 21 |
+
"LABEL_2": 2
|
| 22 |
+
},
|
| 23 |
+
"layer_norm_eps": 1e-12,
|
| 24 |
+
"max_position_embeddings": 512,
|
| 25 |
+
"model_type": "bert",
|
| 26 |
+
"num_attention_heads": 12,
|
| 27 |
+
"num_hidden_layers": 12,
|
| 28 |
+
"output_past": true,
|
| 29 |
+
"pad_token_id": 0,
|
| 30 |
+
"pooler_fc_size": 768,
|
| 31 |
+
"pooler_num_attention_heads": 12,
|
| 32 |
+
"pooler_num_fc_layers": 3,
|
| 33 |
+
"pooler_size_per_head": 128,
|
| 34 |
+
"pooler_type": "first_token_transform",
|
| 35 |
+
"position_embedding_type": "absolute",
|
| 36 |
+
"problem_type": "single_label_classification",
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.50.2",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 29794
|
| 42 |
+
}
|
checkpoint-1610/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:246067edcb15aeca6b6c1fcde5b65a7de1e470723ec8d30533e5cadf8845bbc4
|
| 3 |
+
size 435725300
|
checkpoint-1610/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f23399096e6003e4803f1fe6431a005eb78ad17316c007e452166b315f1e68c8
|
| 3 |
+
size 871566074
|
checkpoint-1610/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:98208890d61b111ad948bcc1b2055f95f8edd40aa2963b0abdd4f4ee0f11deb1
|
| 3 |
+
size 13990
|
checkpoint-1610/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a07bf5254eb0ce278d1e3ac08a67ead9e866fd8aeb27795d69ac84734c0be090
|
| 3 |
+
size 1064
|
checkpoint-1610/trainer_state.json
ADDED
|
@@ -0,0 +1,1194 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1610,
|
| 3 |
+
"best_metric": 0.7374887766651477,
|
| 4 |
+
"best_model_checkpoint": "./results\\checkpoint-1610",
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1610,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.012426219322771047,
|
| 14 |
+
"grad_norm": 7.803389072418213,
|
| 15 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 16 |
+
"loss": 1.0963,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.024852438645542093,
|
| 21 |
+
"grad_norm": 3.206519842147827,
|
| 22 |
+
"learning_rate": 8.000000000000001e-07,
|
| 23 |
+
"loss": 1.0925,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03727865796831314,
|
| 28 |
+
"grad_norm": 6.171759605407715,
|
| 29 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 30 |
+
"loss": 1.082,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.049704877291084186,
|
| 35 |
+
"grad_norm": 5.720030784606934,
|
| 36 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 37 |
+
"loss": 1.0757,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.062131096613855234,
|
| 42 |
+
"grad_norm": 2.8826277256011963,
|
| 43 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 44 |
+
"loss": 1.0982,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.07455731593662628,
|
| 49 |
+
"grad_norm": 3.8596413135528564,
|
| 50 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 51 |
+
"loss": 1.0692,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.08698353525939732,
|
| 56 |
+
"grad_norm": 1.860378623008728,
|
| 57 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 58 |
+
"loss": 1.0862,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.09940975458216837,
|
| 63 |
+
"grad_norm": 1.8722461462020874,
|
| 64 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 65 |
+
"loss": 1.0569,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.11183597390493942,
|
| 70 |
+
"grad_norm": 3.263986349105835,
|
| 71 |
+
"learning_rate": 3.6000000000000003e-06,
|
| 72 |
+
"loss": 1.0725,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.12426219322771047,
|
| 77 |
+
"grad_norm": 2.589475393295288,
|
| 78 |
+
"learning_rate": 4.000000000000001e-06,
|
| 79 |
+
"loss": 1.0447,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1366884125504815,
|
| 84 |
+
"grad_norm": 1.8939751386642456,
|
| 85 |
+
"learning_rate": 4.4e-06,
|
| 86 |
+
"loss": 1.056,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.14911463187325255,
|
| 91 |
+
"grad_norm": 2.082205057144165,
|
| 92 |
+
"learning_rate": 4.800000000000001e-06,
|
| 93 |
+
"loss": 1.0228,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1615408511960236,
|
| 98 |
+
"grad_norm": 3.9843056201934814,
|
| 99 |
+
"learning_rate": 5.2e-06,
|
| 100 |
+
"loss": 1.021,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.17396707051879465,
|
| 105 |
+
"grad_norm": 3.245530128479004,
|
| 106 |
+
"learning_rate": 5.600000000000001e-06,
|
| 107 |
+
"loss": 1.0255,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1863932898415657,
|
| 112 |
+
"grad_norm": 3.017813205718994,
|
| 113 |
+
"learning_rate": 6e-06,
|
| 114 |
+
"loss": 1.016,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.19881950916433674,
|
| 119 |
+
"grad_norm": 3.6596145629882812,
|
| 120 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 121 |
+
"loss": 0.9506,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.2112457284871078,
|
| 126 |
+
"grad_norm": 3.5496115684509277,
|
| 127 |
+
"learning_rate": 6.800000000000001e-06,
|
| 128 |
+
"loss": 0.9531,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.22367194780987884,
|
| 133 |
+
"grad_norm": 7.028264045715332,
|
| 134 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 135 |
+
"loss": 0.9606,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.2360981671326499,
|
| 140 |
+
"grad_norm": 4.832308292388916,
|
| 141 |
+
"learning_rate": 7.600000000000001e-06,
|
| 142 |
+
"loss": 0.9493,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.24852438645542094,
|
| 147 |
+
"grad_norm": 5.981413841247559,
|
| 148 |
+
"learning_rate": 8.000000000000001e-06,
|
| 149 |
+
"loss": 0.9692,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.26095060577819196,
|
| 154 |
+
"grad_norm": 4.8830952644348145,
|
| 155 |
+
"learning_rate": 8.400000000000001e-06,
|
| 156 |
+
"loss": 0.8339,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.273376825100963,
|
| 161 |
+
"grad_norm": 7.470656394958496,
|
| 162 |
+
"learning_rate": 8.8e-06,
|
| 163 |
+
"loss": 0.8902,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.28580304442373405,
|
| 168 |
+
"grad_norm": 10.00381088256836,
|
| 169 |
+
"learning_rate": 9.200000000000002e-06,
|
| 170 |
+
"loss": 0.9514,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2982292637465051,
|
| 175 |
+
"grad_norm": 8.733065605163574,
|
| 176 |
+
"learning_rate": 9.600000000000001e-06,
|
| 177 |
+
"loss": 0.9589,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.31065548306927615,
|
| 182 |
+
"grad_norm": 9.063344955444336,
|
| 183 |
+
"learning_rate": 1e-05,
|
| 184 |
+
"loss": 0.9367,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.3230817023920472,
|
| 189 |
+
"grad_norm": 4.745977401733398,
|
| 190 |
+
"learning_rate": 1.04e-05,
|
| 191 |
+
"loss": 0.8591,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.33550792171481825,
|
| 196 |
+
"grad_norm": 6.759552001953125,
|
| 197 |
+
"learning_rate": 1.0800000000000002e-05,
|
| 198 |
+
"loss": 0.8669,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.3479341410375893,
|
| 203 |
+
"grad_norm": 6.986461639404297,
|
| 204 |
+
"learning_rate": 1.1200000000000001e-05,
|
| 205 |
+
"loss": 0.8973,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.36036036036036034,
|
| 210 |
+
"grad_norm": 9.381807327270508,
|
| 211 |
+
"learning_rate": 1.16e-05,
|
| 212 |
+
"loss": 0.8768,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.3727865796831314,
|
| 217 |
+
"grad_norm": 9.528881072998047,
|
| 218 |
+
"learning_rate": 1.2e-05,
|
| 219 |
+
"loss": 0.9337,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.38521279900590244,
|
| 224 |
+
"grad_norm": 5.089269638061523,
|
| 225 |
+
"learning_rate": 1.2400000000000002e-05,
|
| 226 |
+
"loss": 0.8917,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3976390183286735,
|
| 231 |
+
"grad_norm": 5.263373374938965,
|
| 232 |
+
"learning_rate": 1.2800000000000001e-05,
|
| 233 |
+
"loss": 0.8682,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.41006523765144454,
|
| 238 |
+
"grad_norm": 5.064725399017334,
|
| 239 |
+
"learning_rate": 1.3200000000000002e-05,
|
| 240 |
+
"loss": 0.9005,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.4224914569742156,
|
| 245 |
+
"grad_norm": 7.281830787658691,
|
| 246 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 247 |
+
"loss": 0.83,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.43491767629698663,
|
| 252 |
+
"grad_norm": 6.057513236999512,
|
| 253 |
+
"learning_rate": 1.4e-05,
|
| 254 |
+
"loss": 0.8947,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.4473438956197577,
|
| 259 |
+
"grad_norm": 7.247957229614258,
|
| 260 |
+
"learning_rate": 1.4400000000000001e-05,
|
| 261 |
+
"loss": 0.9045,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.45977011494252873,
|
| 266 |
+
"grad_norm": 5.190288066864014,
|
| 267 |
+
"learning_rate": 1.48e-05,
|
| 268 |
+
"loss": 0.8841,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4721963342652998,
|
| 273 |
+
"grad_norm": 7.859437465667725,
|
| 274 |
+
"learning_rate": 1.5200000000000002e-05,
|
| 275 |
+
"loss": 0.8824,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.4846225535880708,
|
| 280 |
+
"grad_norm": 4.713529586791992,
|
| 281 |
+
"learning_rate": 1.5600000000000003e-05,
|
| 282 |
+
"loss": 0.8779,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.4970487729108419,
|
| 287 |
+
"grad_norm": 5.651578903198242,
|
| 288 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 289 |
+
"loss": 0.868,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.5094749922336129,
|
| 294 |
+
"grad_norm": 8.400832176208496,
|
| 295 |
+
"learning_rate": 1.64e-05,
|
| 296 |
+
"loss": 0.86,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.5219012115563839,
|
| 301 |
+
"grad_norm": 4.4787068367004395,
|
| 302 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 303 |
+
"loss": 0.865,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.534327430879155,
|
| 308 |
+
"grad_norm": 8.719943046569824,
|
| 309 |
+
"learning_rate": 1.72e-05,
|
| 310 |
+
"loss": 0.848,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.546753650201926,
|
| 315 |
+
"grad_norm": 5.370767593383789,
|
| 316 |
+
"learning_rate": 1.76e-05,
|
| 317 |
+
"loss": 0.8235,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.5591798695246971,
|
| 322 |
+
"grad_norm": 7.650132179260254,
|
| 323 |
+
"learning_rate": 1.8e-05,
|
| 324 |
+
"loss": 0.8323,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5716060888474681,
|
| 329 |
+
"grad_norm": 5.548404693603516,
|
| 330 |
+
"learning_rate": 1.8400000000000003e-05,
|
| 331 |
+
"loss": 0.8242,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5840323081702392,
|
| 336 |
+
"grad_norm": 5.651975631713867,
|
| 337 |
+
"learning_rate": 1.88e-05,
|
| 338 |
+
"loss": 0.7487,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5964585274930102,
|
| 343 |
+
"grad_norm": 4.232389450073242,
|
| 344 |
+
"learning_rate": 1.9200000000000003e-05,
|
| 345 |
+
"loss": 0.8622,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.6088847468157813,
|
| 350 |
+
"grad_norm": 5.476607322692871,
|
| 351 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 352 |
+
"loss": 0.8144,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.6213109661385523,
|
| 357 |
+
"grad_norm": 6.335455894470215,
|
| 358 |
+
"learning_rate": 2e-05,
|
| 359 |
+
"loss": 0.8288,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.6337371854613234,
|
| 364 |
+
"grad_norm": 7.2705912590026855,
|
| 365 |
+
"learning_rate": 1.989539748953975e-05,
|
| 366 |
+
"loss": 0.8573,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6461634047840944,
|
| 371 |
+
"grad_norm": 6.81685209274292,
|
| 372 |
+
"learning_rate": 1.97907949790795e-05,
|
| 373 |
+
"loss": 0.8903,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6585896241068655,
|
| 378 |
+
"grad_norm": 5.642577648162842,
|
| 379 |
+
"learning_rate": 1.9686192468619246e-05,
|
| 380 |
+
"loss": 0.8102,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.6710158434296365,
|
| 385 |
+
"grad_norm": 11.23028564453125,
|
| 386 |
+
"learning_rate": 1.9581589958158997e-05,
|
| 387 |
+
"loss": 0.8068,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.6834420627524076,
|
| 392 |
+
"grad_norm": 10.290377616882324,
|
| 393 |
+
"learning_rate": 1.9476987447698744e-05,
|
| 394 |
+
"loss": 0.8092,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.6958682820751786,
|
| 399 |
+
"grad_norm": 7.7588629722595215,
|
| 400 |
+
"learning_rate": 1.9372384937238494e-05,
|
| 401 |
+
"loss": 0.7837,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.7082945013979497,
|
| 406 |
+
"grad_norm": 6.096302509307861,
|
| 407 |
+
"learning_rate": 1.9267782426778245e-05,
|
| 408 |
+
"loss": 0.8138,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.7207207207207207,
|
| 413 |
+
"grad_norm": 9.541234970092773,
|
| 414 |
+
"learning_rate": 1.916317991631799e-05,
|
| 415 |
+
"loss": 0.8009,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.7331469400434918,
|
| 420 |
+
"grad_norm": 6.361993789672852,
|
| 421 |
+
"learning_rate": 1.9058577405857742e-05,
|
| 422 |
+
"loss": 0.8185,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.7455731593662628,
|
| 427 |
+
"grad_norm": 12.062479972839355,
|
| 428 |
+
"learning_rate": 1.8953974895397492e-05,
|
| 429 |
+
"loss": 0.847,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.7579993786890339,
|
| 434 |
+
"grad_norm": 5.53268575668335,
|
| 435 |
+
"learning_rate": 1.884937238493724e-05,
|
| 436 |
+
"loss": 0.7892,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.7704255980118049,
|
| 441 |
+
"grad_norm": 7.868827819824219,
|
| 442 |
+
"learning_rate": 1.874476987447699e-05,
|
| 443 |
+
"loss": 0.7591,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.782851817334576,
|
| 448 |
+
"grad_norm": 7.4785380363464355,
|
| 449 |
+
"learning_rate": 1.8640167364016737e-05,
|
| 450 |
+
"loss": 0.7875,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.795278036657347,
|
| 455 |
+
"grad_norm": 6.592331409454346,
|
| 456 |
+
"learning_rate": 1.8535564853556487e-05,
|
| 457 |
+
"loss": 0.8145,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.8077042559801181,
|
| 462 |
+
"grad_norm": 6.539604663848877,
|
| 463 |
+
"learning_rate": 1.8430962343096234e-05,
|
| 464 |
+
"loss": 0.7509,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.8201304753028891,
|
| 469 |
+
"grad_norm": 8.498805046081543,
|
| 470 |
+
"learning_rate": 1.8326359832635985e-05,
|
| 471 |
+
"loss": 0.8198,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.8325566946256602,
|
| 476 |
+
"grad_norm": 7.241284370422363,
|
| 477 |
+
"learning_rate": 1.8221757322175732e-05,
|
| 478 |
+
"loss": 0.7804,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.8449829139484312,
|
| 483 |
+
"grad_norm": 8.760014533996582,
|
| 484 |
+
"learning_rate": 1.8117154811715482e-05,
|
| 485 |
+
"loss": 0.8414,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.8574091332712023,
|
| 490 |
+
"grad_norm": 4.911290168762207,
|
| 491 |
+
"learning_rate": 1.801255230125523e-05,
|
| 492 |
+
"loss": 0.788,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.8698353525939733,
|
| 497 |
+
"grad_norm": 5.76509952545166,
|
| 498 |
+
"learning_rate": 1.790794979079498e-05,
|
| 499 |
+
"loss": 0.7133,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.8822615719167444,
|
| 504 |
+
"grad_norm": 6.4433088302612305,
|
| 505 |
+
"learning_rate": 1.7803347280334727e-05,
|
| 506 |
+
"loss": 0.7887,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.8946877912395154,
|
| 511 |
+
"grad_norm": 6.976423740386963,
|
| 512 |
+
"learning_rate": 1.7698744769874477e-05,
|
| 513 |
+
"loss": 0.8825,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.9071140105622865,
|
| 518 |
+
"grad_norm": 5.915661811828613,
|
| 519 |
+
"learning_rate": 1.7594142259414228e-05,
|
| 520 |
+
"loss": 0.727,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.9195402298850575,
|
| 525 |
+
"grad_norm": 6.439208030700684,
|
| 526 |
+
"learning_rate": 1.7489539748953975e-05,
|
| 527 |
+
"loss": 0.7372,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.9319664492078286,
|
| 532 |
+
"grad_norm": 7.3419084548950195,
|
| 533 |
+
"learning_rate": 1.7384937238493725e-05,
|
| 534 |
+
"loss": 0.7954,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.9443926685305996,
|
| 539 |
+
"grad_norm": 6.151190280914307,
|
| 540 |
+
"learning_rate": 1.7280334728033476e-05,
|
| 541 |
+
"loss": 0.7261,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.9568188878533707,
|
| 546 |
+
"grad_norm": 6.01263952255249,
|
| 547 |
+
"learning_rate": 1.7175732217573223e-05,
|
| 548 |
+
"loss": 0.773,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.9692451071761417,
|
| 553 |
+
"grad_norm": 6.289830684661865,
|
| 554 |
+
"learning_rate": 1.7071129707112973e-05,
|
| 555 |
+
"loss": 0.8026,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.9816713264989128,
|
| 560 |
+
"grad_norm": 8.667413711547852,
|
| 561 |
+
"learning_rate": 1.696652719665272e-05,
|
| 562 |
+
"loss": 0.7511,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.9940975458216837,
|
| 567 |
+
"grad_norm": 26.640560150146484,
|
| 568 |
+
"learning_rate": 1.686192468619247e-05,
|
| 569 |
+
"loss": 0.7553,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.0,
|
| 574 |
+
"eval_accuracy": 0.6735517937567945,
|
| 575 |
+
"eval_f1": 0.6694435151334517,
|
| 576 |
+
"eval_loss": 0.7382379770278931,
|
| 577 |
+
"eval_precision": 0.6693546214892159,
|
| 578 |
+
"eval_recall": 0.6735517937567945,
|
| 579 |
+
"eval_runtime": 51.9119,
|
| 580 |
+
"eval_samples_per_second": 124.037,
|
| 581 |
+
"eval_steps_per_second": 15.507,
|
| 582 |
+
"step": 805
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 1.0062131096613856,
|
| 586 |
+
"grad_norm": 4.536811828613281,
|
| 587 |
+
"learning_rate": 1.6757322175732218e-05,
|
| 588 |
+
"loss": 0.6675,
|
| 589 |
+
"step": 810
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.0186393289841567,
|
| 593 |
+
"grad_norm": 8.320352554321289,
|
| 594 |
+
"learning_rate": 1.6652719665271968e-05,
|
| 595 |
+
"loss": 0.7224,
|
| 596 |
+
"step": 820
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.0310655483069275,
|
| 600 |
+
"grad_norm": 5.392132759094238,
|
| 601 |
+
"learning_rate": 1.6548117154811715e-05,
|
| 602 |
+
"loss": 0.7079,
|
| 603 |
+
"step": 830
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.0434917676296986,
|
| 607 |
+
"grad_norm": 8.554080963134766,
|
| 608 |
+
"learning_rate": 1.6443514644351466e-05,
|
| 609 |
+
"loss": 0.7505,
|
| 610 |
+
"step": 840
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 1.0559179869524697,
|
| 614 |
+
"grad_norm": 9.378899574279785,
|
| 615 |
+
"learning_rate": 1.6338912133891213e-05,
|
| 616 |
+
"loss": 0.6287,
|
| 617 |
+
"step": 850
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.0683442062752408,
|
| 621 |
+
"grad_norm": 7.220132350921631,
|
| 622 |
+
"learning_rate": 1.6234309623430963e-05,
|
| 623 |
+
"loss": 0.6644,
|
| 624 |
+
"step": 860
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.0807704255980117,
|
| 628 |
+
"grad_norm": 8.708951950073242,
|
| 629 |
+
"learning_rate": 1.612970711297071e-05,
|
| 630 |
+
"loss": 0.7393,
|
| 631 |
+
"step": 870
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 1.0931966449207828,
|
| 635 |
+
"grad_norm": 8.052141189575195,
|
| 636 |
+
"learning_rate": 1.602510460251046e-05,
|
| 637 |
+
"loss": 0.7438,
|
| 638 |
+
"step": 880
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.105622864243554,
|
| 642 |
+
"grad_norm": 7.614938735961914,
|
| 643 |
+
"learning_rate": 1.592050209205021e-05,
|
| 644 |
+
"loss": 0.6631,
|
| 645 |
+
"step": 890
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 1.118049083566325,
|
| 649 |
+
"grad_norm": 10.100054740905762,
|
| 650 |
+
"learning_rate": 1.5815899581589958e-05,
|
| 651 |
+
"loss": 0.5594,
|
| 652 |
+
"step": 900
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 1.130475302889096,
|
| 656 |
+
"grad_norm": 8.444486618041992,
|
| 657 |
+
"learning_rate": 1.571129707112971e-05,
|
| 658 |
+
"loss": 0.6603,
|
| 659 |
+
"step": 910
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.142901522211867,
|
| 663 |
+
"grad_norm": 5.679397106170654,
|
| 664 |
+
"learning_rate": 1.560669456066946e-05,
|
| 665 |
+
"loss": 0.6457,
|
| 666 |
+
"step": 920
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 1.1553277415346381,
|
| 670 |
+
"grad_norm": 5.39376163482666,
|
| 671 |
+
"learning_rate": 1.5502092050209206e-05,
|
| 672 |
+
"loss": 0.6546,
|
| 673 |
+
"step": 930
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 1.167753960857409,
|
| 677 |
+
"grad_norm": 8.007966041564941,
|
| 678 |
+
"learning_rate": 1.5397489539748957e-05,
|
| 679 |
+
"loss": 0.7291,
|
| 680 |
+
"step": 940
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.1801801801801801,
|
| 684 |
+
"grad_norm": 8.760208129882812,
|
| 685 |
+
"learning_rate": 1.5292887029288704e-05,
|
| 686 |
+
"loss": 0.688,
|
| 687 |
+
"step": 950
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 1.1926063995029512,
|
| 691 |
+
"grad_norm": 6.147202491760254,
|
| 692 |
+
"learning_rate": 1.5188284518828454e-05,
|
| 693 |
+
"loss": 0.6608,
|
| 694 |
+
"step": 960
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 1.2050326188257223,
|
| 698 |
+
"grad_norm": 8.280518531799316,
|
| 699 |
+
"learning_rate": 1.5083682008368201e-05,
|
| 700 |
+
"loss": 0.6812,
|
| 701 |
+
"step": 970
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 1.2174588381484934,
|
| 705 |
+
"grad_norm": 5.4499053955078125,
|
| 706 |
+
"learning_rate": 1.4979079497907951e-05,
|
| 707 |
+
"loss": 0.6354,
|
| 708 |
+
"step": 980
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 1.2298850574712643,
|
| 712 |
+
"grad_norm": 9.76456356048584,
|
| 713 |
+
"learning_rate": 1.4874476987447699e-05,
|
| 714 |
+
"loss": 0.6708,
|
| 715 |
+
"step": 990
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 1.2423112767940354,
|
| 719 |
+
"grad_norm": 10.97257137298584,
|
| 720 |
+
"learning_rate": 1.4769874476987449e-05,
|
| 721 |
+
"loss": 0.6937,
|
| 722 |
+
"step": 1000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.2547374961168065,
|
| 726 |
+
"grad_norm": 11.25165843963623,
|
| 727 |
+
"learning_rate": 1.4665271966527198e-05,
|
| 728 |
+
"loss": 0.7248,
|
| 729 |
+
"step": 1010
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.2671637154395774,
|
| 733 |
+
"grad_norm": 8.685588836669922,
|
| 734 |
+
"learning_rate": 1.4560669456066946e-05,
|
| 735 |
+
"loss": 0.7222,
|
| 736 |
+
"step": 1020
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.2795899347623485,
|
| 740 |
+
"grad_norm": 7.983170032501221,
|
| 741 |
+
"learning_rate": 1.4456066945606695e-05,
|
| 742 |
+
"loss": 0.7155,
|
| 743 |
+
"step": 1030
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 1.2920161540851196,
|
| 747 |
+
"grad_norm": 3.756377935409546,
|
| 748 |
+
"learning_rate": 1.4351464435146446e-05,
|
| 749 |
+
"loss": 0.6588,
|
| 750 |
+
"step": 1040
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 1.3044423734078907,
|
| 754 |
+
"grad_norm": 5.793786525726318,
|
| 755 |
+
"learning_rate": 1.4246861924686193e-05,
|
| 756 |
+
"loss": 0.6079,
|
| 757 |
+
"step": 1050
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 1.3168685927306618,
|
| 761 |
+
"grad_norm": 6.8630290031433105,
|
| 762 |
+
"learning_rate": 1.4142259414225943e-05,
|
| 763 |
+
"loss": 0.6038,
|
| 764 |
+
"step": 1060
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 1.3292948120534327,
|
| 768 |
+
"grad_norm": 7.02629280090332,
|
| 769 |
+
"learning_rate": 1.403765690376569e-05,
|
| 770 |
+
"loss": 0.6587,
|
| 771 |
+
"step": 1070
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 1.3417210313762038,
|
| 775 |
+
"grad_norm": 5.012918949127197,
|
| 776 |
+
"learning_rate": 1.393305439330544e-05,
|
| 777 |
+
"loss": 0.6479,
|
| 778 |
+
"step": 1080
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 1.354147250698975,
|
| 782 |
+
"grad_norm": 9.064984321594238,
|
| 783 |
+
"learning_rate": 1.382845188284519e-05,
|
| 784 |
+
"loss": 0.6771,
|
| 785 |
+
"step": 1090
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 1.3665734700217458,
|
| 789 |
+
"grad_norm": 7.3286638259887695,
|
| 790 |
+
"learning_rate": 1.3723849372384938e-05,
|
| 791 |
+
"loss": 0.6645,
|
| 792 |
+
"step": 1100
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 1.3789996893445169,
|
| 796 |
+
"grad_norm": 10.485735893249512,
|
| 797 |
+
"learning_rate": 1.3619246861924687e-05,
|
| 798 |
+
"loss": 0.7744,
|
| 799 |
+
"step": 1110
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 1.391425908667288,
|
| 803 |
+
"grad_norm": 8.652807235717773,
|
| 804 |
+
"learning_rate": 1.3514644351464437e-05,
|
| 805 |
+
"loss": 0.6424,
|
| 806 |
+
"step": 1120
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 1.403852127990059,
|
| 810 |
+
"grad_norm": 11.628154754638672,
|
| 811 |
+
"learning_rate": 1.3410041841004184e-05,
|
| 812 |
+
"loss": 0.7039,
|
| 813 |
+
"step": 1130
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 1.4162783473128302,
|
| 817 |
+
"grad_norm": 10.061933517456055,
|
| 818 |
+
"learning_rate": 1.3305439330543935e-05,
|
| 819 |
+
"loss": 0.676,
|
| 820 |
+
"step": 1140
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 1.428704566635601,
|
| 824 |
+
"grad_norm": 5.85321044921875,
|
| 825 |
+
"learning_rate": 1.3200836820083682e-05,
|
| 826 |
+
"loss": 0.6032,
|
| 827 |
+
"step": 1150
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.4411307859583722,
|
| 831 |
+
"grad_norm": 5.4685378074646,
|
| 832 |
+
"learning_rate": 1.3096234309623432e-05,
|
| 833 |
+
"loss": 0.6062,
|
| 834 |
+
"step": 1160
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.4535570052811433,
|
| 838 |
+
"grad_norm": 7.55083703994751,
|
| 839 |
+
"learning_rate": 1.2991631799163181e-05,
|
| 840 |
+
"loss": 0.6231,
|
| 841 |
+
"step": 1170
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.4659832246039142,
|
| 845 |
+
"grad_norm": 9.285325050354004,
|
| 846 |
+
"learning_rate": 1.288702928870293e-05,
|
| 847 |
+
"loss": 0.6233,
|
| 848 |
+
"step": 1180
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.4784094439266853,
|
| 852 |
+
"grad_norm": 8.982355117797852,
|
| 853 |
+
"learning_rate": 1.2782426778242678e-05,
|
| 854 |
+
"loss": 0.5726,
|
| 855 |
+
"step": 1190
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 1.4908356632494564,
|
| 859 |
+
"grad_norm": 17.36650848388672,
|
| 860 |
+
"learning_rate": 1.2677824267782429e-05,
|
| 861 |
+
"loss": 0.5925,
|
| 862 |
+
"step": 1200
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 1.5032618825722275,
|
| 866 |
+
"grad_norm": 6.095853805541992,
|
| 867 |
+
"learning_rate": 1.2573221757322176e-05,
|
| 868 |
+
"loss": 0.6963,
|
| 869 |
+
"step": 1210
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.5156881018949986,
|
| 873 |
+
"grad_norm": 4.788658618927002,
|
| 874 |
+
"learning_rate": 1.2468619246861926e-05,
|
| 875 |
+
"loss": 0.6042,
|
| 876 |
+
"step": 1220
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.5281143212177695,
|
| 880 |
+
"grad_norm": 15.81484317779541,
|
| 881 |
+
"learning_rate": 1.2364016736401673e-05,
|
| 882 |
+
"loss": 0.6201,
|
| 883 |
+
"step": 1230
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.5405405405405406,
|
| 887 |
+
"grad_norm": 8.392333030700684,
|
| 888 |
+
"learning_rate": 1.2259414225941424e-05,
|
| 889 |
+
"loss": 0.718,
|
| 890 |
+
"step": 1240
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.5529667598633115,
|
| 894 |
+
"grad_norm": 9.427897453308105,
|
| 895 |
+
"learning_rate": 1.2154811715481171e-05,
|
| 896 |
+
"loss": 0.6265,
|
| 897 |
+
"step": 1250
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.5653929791860826,
|
| 901 |
+
"grad_norm": 5.241062641143799,
|
| 902 |
+
"learning_rate": 1.2050209205020921e-05,
|
| 903 |
+
"loss": 0.6071,
|
| 904 |
+
"step": 1260
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.5778191985088537,
|
| 908 |
+
"grad_norm": 7.011557579040527,
|
| 909 |
+
"learning_rate": 1.194560669456067e-05,
|
| 910 |
+
"loss": 0.6251,
|
| 911 |
+
"step": 1270
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.5902454178316248,
|
| 915 |
+
"grad_norm": 5.581645965576172,
|
| 916 |
+
"learning_rate": 1.184100418410042e-05,
|
| 917 |
+
"loss": 0.6214,
|
| 918 |
+
"step": 1280
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 1.6026716371543959,
|
| 922 |
+
"grad_norm": 11.603231430053711,
|
| 923 |
+
"learning_rate": 1.1736401673640168e-05,
|
| 924 |
+
"loss": 0.6024,
|
| 925 |
+
"step": 1290
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 1.615097856477167,
|
| 929 |
+
"grad_norm": 9.178550720214844,
|
| 930 |
+
"learning_rate": 1.1631799163179918e-05,
|
| 931 |
+
"loss": 0.6294,
|
| 932 |
+
"step": 1300
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.6275240757999379,
|
| 936 |
+
"grad_norm": 5.397329330444336,
|
| 937 |
+
"learning_rate": 1.1527196652719665e-05,
|
| 938 |
+
"loss": 0.6429,
|
| 939 |
+
"step": 1310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.639950295122709,
|
| 943 |
+
"grad_norm": 6.596381664276123,
|
| 944 |
+
"learning_rate": 1.1422594142259416e-05,
|
| 945 |
+
"loss": 0.6326,
|
| 946 |
+
"step": 1320
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.6523765144454798,
|
| 950 |
+
"grad_norm": 10.257133483886719,
|
| 951 |
+
"learning_rate": 1.1317991631799163e-05,
|
| 952 |
+
"loss": 0.626,
|
| 953 |
+
"step": 1330
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 1.664802733768251,
|
| 957 |
+
"grad_norm": 9.520708084106445,
|
| 958 |
+
"learning_rate": 1.1213389121338913e-05,
|
| 959 |
+
"loss": 0.6784,
|
| 960 |
+
"step": 1340
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 1.677228953091022,
|
| 964 |
+
"grad_norm": 8.193737030029297,
|
| 965 |
+
"learning_rate": 1.1108786610878662e-05,
|
| 966 |
+
"loss": 0.67,
|
| 967 |
+
"step": 1350
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 1.6896551724137931,
|
| 971 |
+
"grad_norm": 10.557798385620117,
|
| 972 |
+
"learning_rate": 1.100418410041841e-05,
|
| 973 |
+
"loss": 0.6265,
|
| 974 |
+
"step": 1360
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.7020813917365643,
|
| 978 |
+
"grad_norm": 5.501838684082031,
|
| 979 |
+
"learning_rate": 1.089958158995816e-05,
|
| 980 |
+
"loss": 0.6588,
|
| 981 |
+
"step": 1370
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 1.7145076110593354,
|
| 985 |
+
"grad_norm": 9.010986328125,
|
| 986 |
+
"learning_rate": 1.079497907949791e-05,
|
| 987 |
+
"loss": 0.6238,
|
| 988 |
+
"step": 1380
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.7269338303821062,
|
| 992 |
+
"grad_norm": 8.921228408813477,
|
| 993 |
+
"learning_rate": 1.0690376569037657e-05,
|
| 994 |
+
"loss": 0.669,
|
| 995 |
+
"step": 1390
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 1.7393600497048773,
|
| 999 |
+
"grad_norm": 8.683089256286621,
|
| 1000 |
+
"learning_rate": 1.0585774058577407e-05,
|
| 1001 |
+
"loss": 0.5655,
|
| 1002 |
+
"step": 1400
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 1.7517862690276482,
|
| 1006 |
+
"grad_norm": 11.0118408203125,
|
| 1007 |
+
"learning_rate": 1.0481171548117154e-05,
|
| 1008 |
+
"loss": 0.6968,
|
| 1009 |
+
"step": 1410
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.7642124883504193,
|
| 1013 |
+
"grad_norm": 7.634212970733643,
|
| 1014 |
+
"learning_rate": 1.0376569037656905e-05,
|
| 1015 |
+
"loss": 0.5995,
|
| 1016 |
+
"step": 1420
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.7766387076731904,
|
| 1020 |
+
"grad_norm": 7.622481346130371,
|
| 1021 |
+
"learning_rate": 1.0271966527196653e-05,
|
| 1022 |
+
"loss": 0.67,
|
| 1023 |
+
"step": 1430
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 1.7890649269959615,
|
| 1027 |
+
"grad_norm": 9.09194278717041,
|
| 1028 |
+
"learning_rate": 1.0167364016736402e-05,
|
| 1029 |
+
"loss": 0.6026,
|
| 1030 |
+
"step": 1440
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 1.8014911463187326,
|
| 1034 |
+
"grad_norm": 8.283378601074219,
|
| 1035 |
+
"learning_rate": 1.0062761506276151e-05,
|
| 1036 |
+
"loss": 0.6119,
|
| 1037 |
+
"step": 1450
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 1.8139173656415035,
|
| 1041 |
+
"grad_norm": 6.903816223144531,
|
| 1042 |
+
"learning_rate": 9.958158995815901e-06,
|
| 1043 |
+
"loss": 0.6326,
|
| 1044 |
+
"step": 1460
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.8263435849642746,
|
| 1048 |
+
"grad_norm": 6.750952243804932,
|
| 1049 |
+
"learning_rate": 9.85355648535565e-06,
|
| 1050 |
+
"loss": 0.6707,
|
| 1051 |
+
"step": 1470
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 1.8387698042870455,
|
| 1055 |
+
"grad_norm": 7.868640899658203,
|
| 1056 |
+
"learning_rate": 9.748953974895399e-06,
|
| 1057 |
+
"loss": 0.6488,
|
| 1058 |
+
"step": 1480
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 1.8511960236098166,
|
| 1062 |
+
"grad_norm": 7.722754001617432,
|
| 1063 |
+
"learning_rate": 9.644351464435148e-06,
|
| 1064 |
+
"loss": 0.6477,
|
| 1065 |
+
"step": 1490
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 1.8636222429325877,
|
| 1069 |
+
"grad_norm": 5.540498733520508,
|
| 1070 |
+
"learning_rate": 9.539748953974896e-06,
|
| 1071 |
+
"loss": 0.6674,
|
| 1072 |
+
"step": 1500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 1.8760484622553588,
|
| 1076 |
+
"grad_norm": 8.326912879943848,
|
| 1077 |
+
"learning_rate": 9.435146443514645e-06,
|
| 1078 |
+
"loss": 0.5612,
|
| 1079 |
+
"step": 1510
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 1.88847468157813,
|
| 1083 |
+
"grad_norm": 10.114307403564453,
|
| 1084 |
+
"learning_rate": 9.330543933054394e-06,
|
| 1085 |
+
"loss": 0.6207,
|
| 1086 |
+
"step": 1520
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 1.900900900900901,
|
| 1090 |
+
"grad_norm": 9.120221138000488,
|
| 1091 |
+
"learning_rate": 9.225941422594144e-06,
|
| 1092 |
+
"loss": 0.5838,
|
| 1093 |
+
"step": 1530
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 1.913327120223672,
|
| 1097 |
+
"grad_norm": 9.336804389953613,
|
| 1098 |
+
"learning_rate": 9.121338912133893e-06,
|
| 1099 |
+
"loss": 0.5486,
|
| 1100 |
+
"step": 1540
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 1.925753339546443,
|
| 1104 |
+
"grad_norm": 6.764523506164551,
|
| 1105 |
+
"learning_rate": 9.016736401673642e-06,
|
| 1106 |
+
"loss": 0.6323,
|
| 1107 |
+
"step": 1550
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 1.938179558869214,
|
| 1111 |
+
"grad_norm": 9.537332534790039,
|
| 1112 |
+
"learning_rate": 8.91213389121339e-06,
|
| 1113 |
+
"loss": 0.5779,
|
| 1114 |
+
"step": 1560
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 1.950605778191985,
|
| 1118 |
+
"grad_norm": 7.732354164123535,
|
| 1119 |
+
"learning_rate": 8.80753138075314e-06,
|
| 1120 |
+
"loss": 0.6373,
|
| 1121 |
+
"step": 1570
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 1.963031997514756,
|
| 1125 |
+
"grad_norm": 10.750661849975586,
|
| 1126 |
+
"learning_rate": 8.702928870292888e-06,
|
| 1127 |
+
"loss": 0.662,
|
| 1128 |
+
"step": 1580
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.9754582168375272,
|
| 1132 |
+
"grad_norm": 5.610159397125244,
|
| 1133 |
+
"learning_rate": 8.598326359832637e-06,
|
| 1134 |
+
"loss": 0.6219,
|
| 1135 |
+
"step": 1590
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 1.9878844361602983,
|
| 1139 |
+
"grad_norm": 5.781522274017334,
|
| 1140 |
+
"learning_rate": 8.493723849372385e-06,
|
| 1141 |
+
"loss": 0.5479,
|
| 1142 |
+
"step": 1600
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.0,
|
| 1146 |
+
"grad_norm": 4.716365814208984,
|
| 1147 |
+
"learning_rate": 8.389121338912136e-06,
|
| 1148 |
+
"loss": 0.5645,
|
| 1149 |
+
"step": 1610
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.0,
|
| 1153 |
+
"eval_accuracy": 0.7378474918465601,
|
| 1154 |
+
"eval_f1": 0.7374887766651477,
|
| 1155 |
+
"eval_loss": 0.6107740998268127,
|
| 1156 |
+
"eval_precision": 0.7408097797843906,
|
| 1157 |
+
"eval_recall": 0.7378474918465601,
|
| 1158 |
+
"eval_runtime": 51.7148,
|
| 1159 |
+
"eval_samples_per_second": 124.51,
|
| 1160 |
+
"eval_steps_per_second": 15.566,
|
| 1161 |
+
"step": 1610
|
| 1162 |
+
}
|
| 1163 |
+
],
|
| 1164 |
+
"logging_steps": 10,
|
| 1165 |
+
"max_steps": 2412,
|
| 1166 |
+
"num_input_tokens_seen": 0,
|
| 1167 |
+
"num_train_epochs": 3,
|
| 1168 |
+
"save_steps": 500,
|
| 1169 |
+
"stateful_callbacks": {
|
| 1170 |
+
"EarlyStoppingCallback": {
|
| 1171 |
+
"args": {
|
| 1172 |
+
"early_stopping_patience": 3,
|
| 1173 |
+
"early_stopping_threshold": 0.0
|
| 1174 |
+
},
|
| 1175 |
+
"attributes": {
|
| 1176 |
+
"early_stopping_patience_counter": 0
|
| 1177 |
+
}
|
| 1178 |
+
},
|
| 1179 |
+
"TrainerControl": {
|
| 1180 |
+
"args": {
|
| 1181 |
+
"should_epoch_stop": false,
|
| 1182 |
+
"should_evaluate": false,
|
| 1183 |
+
"should_log": false,
|
| 1184 |
+
"should_save": true,
|
| 1185 |
+
"should_training_stop": false
|
| 1186 |
+
},
|
| 1187 |
+
"attributes": {}
|
| 1188 |
+
}
|
| 1189 |
+
},
|
| 1190 |
+
"total_flos": 370545915104064.0,
|
| 1191 |
+
"train_batch_size": 8,
|
| 1192 |
+
"trial_name": null,
|
| 1193 |
+
"trial_params": null
|
| 1194 |
+
}
|
checkpoint-1610/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b74a6de7f3ed0719b44759cf4f2718a423a057052a06b70b41090640f4b42b14
|
| 3 |
+
size 5304
|
checkpoint-2412/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"directionality": "bidi",
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"LABEL_0": 0,
|
| 20 |
+
"LABEL_1": 1,
|
| 21 |
+
"LABEL_2": 2
|
| 22 |
+
},
|
| 23 |
+
"layer_norm_eps": 1e-12,
|
| 24 |
+
"max_position_embeddings": 512,
|
| 25 |
+
"model_type": "bert",
|
| 26 |
+
"num_attention_heads": 12,
|
| 27 |
+
"num_hidden_layers": 12,
|
| 28 |
+
"output_past": true,
|
| 29 |
+
"pad_token_id": 0,
|
| 30 |
+
"pooler_fc_size": 768,
|
| 31 |
+
"pooler_num_attention_heads": 12,
|
| 32 |
+
"pooler_num_fc_layers": 3,
|
| 33 |
+
"pooler_size_per_head": 128,
|
| 34 |
+
"pooler_type": "first_token_transform",
|
| 35 |
+
"position_embedding_type": "absolute",
|
| 36 |
+
"problem_type": "single_label_classification",
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.50.2",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 29794
|
| 42 |
+
}
|
checkpoint-2412/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5e2549dabd16a904f8d8e2a6da81bbfd7d4245fca26114dee7a649068123d9f
|
| 3 |
+
size 435725300
|
checkpoint-2412/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61d40480040a84925019769c704e00f860024d70ba44734a779bcbf586bf2706
|
| 3 |
+
size 871566074
|
checkpoint-2412/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6468c28627d0b79a9dafeeaf54c6b7a13fd210a8759313da4bf978082c4d65e5
|
| 3 |
+
size 13990
|
checkpoint-2412/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cabb011bd12ccf9e32529cc09805490738127f1375ee9cfa45e7f3761b2a3eee
|
| 3 |
+
size 1064
|
checkpoint-2412/trainer_state.json
ADDED
|
@@ -0,0 +1,1766 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 2412,
|
| 3 |
+
"best_metric": 0.766151133896168,
|
| 4 |
+
"best_model_checkpoint": "./results\\checkpoint-2412",
|
| 5 |
+
"epoch": 2.996582789686238,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2412,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.012426219322771047,
|
| 14 |
+
"grad_norm": 7.803389072418213,
|
| 15 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 16 |
+
"loss": 1.0963,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.024852438645542093,
|
| 21 |
+
"grad_norm": 3.206519842147827,
|
| 22 |
+
"learning_rate": 8.000000000000001e-07,
|
| 23 |
+
"loss": 1.0925,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03727865796831314,
|
| 28 |
+
"grad_norm": 6.171759605407715,
|
| 29 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 30 |
+
"loss": 1.082,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.049704877291084186,
|
| 35 |
+
"grad_norm": 5.720030784606934,
|
| 36 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 37 |
+
"loss": 1.0757,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.062131096613855234,
|
| 42 |
+
"grad_norm": 2.8826277256011963,
|
| 43 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 44 |
+
"loss": 1.0982,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.07455731593662628,
|
| 49 |
+
"grad_norm": 3.8596413135528564,
|
| 50 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 51 |
+
"loss": 1.0692,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.08698353525939732,
|
| 56 |
+
"grad_norm": 1.860378623008728,
|
| 57 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 58 |
+
"loss": 1.0862,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.09940975458216837,
|
| 63 |
+
"grad_norm": 1.8722461462020874,
|
| 64 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 65 |
+
"loss": 1.0569,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.11183597390493942,
|
| 70 |
+
"grad_norm": 3.263986349105835,
|
| 71 |
+
"learning_rate": 3.6000000000000003e-06,
|
| 72 |
+
"loss": 1.0725,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.12426219322771047,
|
| 77 |
+
"grad_norm": 2.589475393295288,
|
| 78 |
+
"learning_rate": 4.000000000000001e-06,
|
| 79 |
+
"loss": 1.0447,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1366884125504815,
|
| 84 |
+
"grad_norm": 1.8939751386642456,
|
| 85 |
+
"learning_rate": 4.4e-06,
|
| 86 |
+
"loss": 1.056,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.14911463187325255,
|
| 91 |
+
"grad_norm": 2.082205057144165,
|
| 92 |
+
"learning_rate": 4.800000000000001e-06,
|
| 93 |
+
"loss": 1.0228,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1615408511960236,
|
| 98 |
+
"grad_norm": 3.9843056201934814,
|
| 99 |
+
"learning_rate": 5.2e-06,
|
| 100 |
+
"loss": 1.021,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.17396707051879465,
|
| 105 |
+
"grad_norm": 3.245530128479004,
|
| 106 |
+
"learning_rate": 5.600000000000001e-06,
|
| 107 |
+
"loss": 1.0255,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1863932898415657,
|
| 112 |
+
"grad_norm": 3.017813205718994,
|
| 113 |
+
"learning_rate": 6e-06,
|
| 114 |
+
"loss": 1.016,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.19881950916433674,
|
| 119 |
+
"grad_norm": 3.6596145629882812,
|
| 120 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 121 |
+
"loss": 0.9506,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.2112457284871078,
|
| 126 |
+
"grad_norm": 3.5496115684509277,
|
| 127 |
+
"learning_rate": 6.800000000000001e-06,
|
| 128 |
+
"loss": 0.9531,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.22367194780987884,
|
| 133 |
+
"grad_norm": 7.028264045715332,
|
| 134 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 135 |
+
"loss": 0.9606,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.2360981671326499,
|
| 140 |
+
"grad_norm": 4.832308292388916,
|
| 141 |
+
"learning_rate": 7.600000000000001e-06,
|
| 142 |
+
"loss": 0.9493,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.24852438645542094,
|
| 147 |
+
"grad_norm": 5.981413841247559,
|
| 148 |
+
"learning_rate": 8.000000000000001e-06,
|
| 149 |
+
"loss": 0.9692,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.26095060577819196,
|
| 154 |
+
"grad_norm": 4.8830952644348145,
|
| 155 |
+
"learning_rate": 8.400000000000001e-06,
|
| 156 |
+
"loss": 0.8339,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.273376825100963,
|
| 161 |
+
"grad_norm": 7.470656394958496,
|
| 162 |
+
"learning_rate": 8.8e-06,
|
| 163 |
+
"loss": 0.8902,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.28580304442373405,
|
| 168 |
+
"grad_norm": 10.00381088256836,
|
| 169 |
+
"learning_rate": 9.200000000000002e-06,
|
| 170 |
+
"loss": 0.9514,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2982292637465051,
|
| 175 |
+
"grad_norm": 8.733065605163574,
|
| 176 |
+
"learning_rate": 9.600000000000001e-06,
|
| 177 |
+
"loss": 0.9589,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.31065548306927615,
|
| 182 |
+
"grad_norm": 9.063344955444336,
|
| 183 |
+
"learning_rate": 1e-05,
|
| 184 |
+
"loss": 0.9367,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.3230817023920472,
|
| 189 |
+
"grad_norm": 4.745977401733398,
|
| 190 |
+
"learning_rate": 1.04e-05,
|
| 191 |
+
"loss": 0.8591,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.33550792171481825,
|
| 196 |
+
"grad_norm": 6.759552001953125,
|
| 197 |
+
"learning_rate": 1.0800000000000002e-05,
|
| 198 |
+
"loss": 0.8669,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.3479341410375893,
|
| 203 |
+
"grad_norm": 6.986461639404297,
|
| 204 |
+
"learning_rate": 1.1200000000000001e-05,
|
| 205 |
+
"loss": 0.8973,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.36036036036036034,
|
| 210 |
+
"grad_norm": 9.381807327270508,
|
| 211 |
+
"learning_rate": 1.16e-05,
|
| 212 |
+
"loss": 0.8768,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.3727865796831314,
|
| 217 |
+
"grad_norm": 9.528881072998047,
|
| 218 |
+
"learning_rate": 1.2e-05,
|
| 219 |
+
"loss": 0.9337,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.38521279900590244,
|
| 224 |
+
"grad_norm": 5.089269638061523,
|
| 225 |
+
"learning_rate": 1.2400000000000002e-05,
|
| 226 |
+
"loss": 0.8917,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3976390183286735,
|
| 231 |
+
"grad_norm": 5.263373374938965,
|
| 232 |
+
"learning_rate": 1.2800000000000001e-05,
|
| 233 |
+
"loss": 0.8682,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.41006523765144454,
|
| 238 |
+
"grad_norm": 5.064725399017334,
|
| 239 |
+
"learning_rate": 1.3200000000000002e-05,
|
| 240 |
+
"loss": 0.9005,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.4224914569742156,
|
| 245 |
+
"grad_norm": 7.281830787658691,
|
| 246 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 247 |
+
"loss": 0.83,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.43491767629698663,
|
| 252 |
+
"grad_norm": 6.057513236999512,
|
| 253 |
+
"learning_rate": 1.4e-05,
|
| 254 |
+
"loss": 0.8947,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.4473438956197577,
|
| 259 |
+
"grad_norm": 7.247957229614258,
|
| 260 |
+
"learning_rate": 1.4400000000000001e-05,
|
| 261 |
+
"loss": 0.9045,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.45977011494252873,
|
| 266 |
+
"grad_norm": 5.190288066864014,
|
| 267 |
+
"learning_rate": 1.48e-05,
|
| 268 |
+
"loss": 0.8841,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4721963342652998,
|
| 273 |
+
"grad_norm": 7.859437465667725,
|
| 274 |
+
"learning_rate": 1.5200000000000002e-05,
|
| 275 |
+
"loss": 0.8824,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.4846225535880708,
|
| 280 |
+
"grad_norm": 4.713529586791992,
|
| 281 |
+
"learning_rate": 1.5600000000000003e-05,
|
| 282 |
+
"loss": 0.8779,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.4970487729108419,
|
| 287 |
+
"grad_norm": 5.651578903198242,
|
| 288 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 289 |
+
"loss": 0.868,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.5094749922336129,
|
| 294 |
+
"grad_norm": 8.400832176208496,
|
| 295 |
+
"learning_rate": 1.64e-05,
|
| 296 |
+
"loss": 0.86,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.5219012115563839,
|
| 301 |
+
"grad_norm": 4.4787068367004395,
|
| 302 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 303 |
+
"loss": 0.865,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.534327430879155,
|
| 308 |
+
"grad_norm": 8.719943046569824,
|
| 309 |
+
"learning_rate": 1.72e-05,
|
| 310 |
+
"loss": 0.848,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.546753650201926,
|
| 315 |
+
"grad_norm": 5.370767593383789,
|
| 316 |
+
"learning_rate": 1.76e-05,
|
| 317 |
+
"loss": 0.8235,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.5591798695246971,
|
| 322 |
+
"grad_norm": 7.650132179260254,
|
| 323 |
+
"learning_rate": 1.8e-05,
|
| 324 |
+
"loss": 0.8323,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5716060888474681,
|
| 329 |
+
"grad_norm": 5.548404693603516,
|
| 330 |
+
"learning_rate": 1.8400000000000003e-05,
|
| 331 |
+
"loss": 0.8242,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5840323081702392,
|
| 336 |
+
"grad_norm": 5.651975631713867,
|
| 337 |
+
"learning_rate": 1.88e-05,
|
| 338 |
+
"loss": 0.7487,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5964585274930102,
|
| 343 |
+
"grad_norm": 4.232389450073242,
|
| 344 |
+
"learning_rate": 1.9200000000000003e-05,
|
| 345 |
+
"loss": 0.8622,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.6088847468157813,
|
| 350 |
+
"grad_norm": 5.476607322692871,
|
| 351 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 352 |
+
"loss": 0.8144,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.6213109661385523,
|
| 357 |
+
"grad_norm": 6.335455894470215,
|
| 358 |
+
"learning_rate": 2e-05,
|
| 359 |
+
"loss": 0.8288,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.6337371854613234,
|
| 364 |
+
"grad_norm": 7.2705912590026855,
|
| 365 |
+
"learning_rate": 1.989539748953975e-05,
|
| 366 |
+
"loss": 0.8573,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6461634047840944,
|
| 371 |
+
"grad_norm": 6.81685209274292,
|
| 372 |
+
"learning_rate": 1.97907949790795e-05,
|
| 373 |
+
"loss": 0.8903,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6585896241068655,
|
| 378 |
+
"grad_norm": 5.642577648162842,
|
| 379 |
+
"learning_rate": 1.9686192468619246e-05,
|
| 380 |
+
"loss": 0.8102,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.6710158434296365,
|
| 385 |
+
"grad_norm": 11.23028564453125,
|
| 386 |
+
"learning_rate": 1.9581589958158997e-05,
|
| 387 |
+
"loss": 0.8068,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.6834420627524076,
|
| 392 |
+
"grad_norm": 10.290377616882324,
|
| 393 |
+
"learning_rate": 1.9476987447698744e-05,
|
| 394 |
+
"loss": 0.8092,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.6958682820751786,
|
| 399 |
+
"grad_norm": 7.7588629722595215,
|
| 400 |
+
"learning_rate": 1.9372384937238494e-05,
|
| 401 |
+
"loss": 0.7837,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.7082945013979497,
|
| 406 |
+
"grad_norm": 6.096302509307861,
|
| 407 |
+
"learning_rate": 1.9267782426778245e-05,
|
| 408 |
+
"loss": 0.8138,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.7207207207207207,
|
| 413 |
+
"grad_norm": 9.541234970092773,
|
| 414 |
+
"learning_rate": 1.916317991631799e-05,
|
| 415 |
+
"loss": 0.8009,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.7331469400434918,
|
| 420 |
+
"grad_norm": 6.361993789672852,
|
| 421 |
+
"learning_rate": 1.9058577405857742e-05,
|
| 422 |
+
"loss": 0.8185,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.7455731593662628,
|
| 427 |
+
"grad_norm": 12.062479972839355,
|
| 428 |
+
"learning_rate": 1.8953974895397492e-05,
|
| 429 |
+
"loss": 0.847,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.7579993786890339,
|
| 434 |
+
"grad_norm": 5.53268575668335,
|
| 435 |
+
"learning_rate": 1.884937238493724e-05,
|
| 436 |
+
"loss": 0.7892,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.7704255980118049,
|
| 441 |
+
"grad_norm": 7.868827819824219,
|
| 442 |
+
"learning_rate": 1.874476987447699e-05,
|
| 443 |
+
"loss": 0.7591,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.782851817334576,
|
| 448 |
+
"grad_norm": 7.4785380363464355,
|
| 449 |
+
"learning_rate": 1.8640167364016737e-05,
|
| 450 |
+
"loss": 0.7875,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.795278036657347,
|
| 455 |
+
"grad_norm": 6.592331409454346,
|
| 456 |
+
"learning_rate": 1.8535564853556487e-05,
|
| 457 |
+
"loss": 0.8145,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.8077042559801181,
|
| 462 |
+
"grad_norm": 6.539604663848877,
|
| 463 |
+
"learning_rate": 1.8430962343096234e-05,
|
| 464 |
+
"loss": 0.7509,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.8201304753028891,
|
| 469 |
+
"grad_norm": 8.498805046081543,
|
| 470 |
+
"learning_rate": 1.8326359832635985e-05,
|
| 471 |
+
"loss": 0.8198,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.8325566946256602,
|
| 476 |
+
"grad_norm": 7.241284370422363,
|
| 477 |
+
"learning_rate": 1.8221757322175732e-05,
|
| 478 |
+
"loss": 0.7804,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.8449829139484312,
|
| 483 |
+
"grad_norm": 8.760014533996582,
|
| 484 |
+
"learning_rate": 1.8117154811715482e-05,
|
| 485 |
+
"loss": 0.8414,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.8574091332712023,
|
| 490 |
+
"grad_norm": 4.911290168762207,
|
| 491 |
+
"learning_rate": 1.801255230125523e-05,
|
| 492 |
+
"loss": 0.788,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.8698353525939733,
|
| 497 |
+
"grad_norm": 5.76509952545166,
|
| 498 |
+
"learning_rate": 1.790794979079498e-05,
|
| 499 |
+
"loss": 0.7133,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.8822615719167444,
|
| 504 |
+
"grad_norm": 6.4433088302612305,
|
| 505 |
+
"learning_rate": 1.7803347280334727e-05,
|
| 506 |
+
"loss": 0.7887,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.8946877912395154,
|
| 511 |
+
"grad_norm": 6.976423740386963,
|
| 512 |
+
"learning_rate": 1.7698744769874477e-05,
|
| 513 |
+
"loss": 0.8825,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.9071140105622865,
|
| 518 |
+
"grad_norm": 5.915661811828613,
|
| 519 |
+
"learning_rate": 1.7594142259414228e-05,
|
| 520 |
+
"loss": 0.727,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.9195402298850575,
|
| 525 |
+
"grad_norm": 6.439208030700684,
|
| 526 |
+
"learning_rate": 1.7489539748953975e-05,
|
| 527 |
+
"loss": 0.7372,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.9319664492078286,
|
| 532 |
+
"grad_norm": 7.3419084548950195,
|
| 533 |
+
"learning_rate": 1.7384937238493725e-05,
|
| 534 |
+
"loss": 0.7954,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.9443926685305996,
|
| 539 |
+
"grad_norm": 6.151190280914307,
|
| 540 |
+
"learning_rate": 1.7280334728033476e-05,
|
| 541 |
+
"loss": 0.7261,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.9568188878533707,
|
| 546 |
+
"grad_norm": 6.01263952255249,
|
| 547 |
+
"learning_rate": 1.7175732217573223e-05,
|
| 548 |
+
"loss": 0.773,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.9692451071761417,
|
| 553 |
+
"grad_norm": 6.289830684661865,
|
| 554 |
+
"learning_rate": 1.7071129707112973e-05,
|
| 555 |
+
"loss": 0.8026,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.9816713264989128,
|
| 560 |
+
"grad_norm": 8.667413711547852,
|
| 561 |
+
"learning_rate": 1.696652719665272e-05,
|
| 562 |
+
"loss": 0.7511,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.9940975458216837,
|
| 567 |
+
"grad_norm": 26.640560150146484,
|
| 568 |
+
"learning_rate": 1.686192468619247e-05,
|
| 569 |
+
"loss": 0.7553,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.0,
|
| 574 |
+
"eval_accuracy": 0.6735517937567945,
|
| 575 |
+
"eval_f1": 0.6694435151334517,
|
| 576 |
+
"eval_loss": 0.7382379770278931,
|
| 577 |
+
"eval_precision": 0.6693546214892159,
|
| 578 |
+
"eval_recall": 0.6735517937567945,
|
| 579 |
+
"eval_runtime": 51.9119,
|
| 580 |
+
"eval_samples_per_second": 124.037,
|
| 581 |
+
"eval_steps_per_second": 15.507,
|
| 582 |
+
"step": 805
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 1.0062131096613856,
|
| 586 |
+
"grad_norm": 4.536811828613281,
|
| 587 |
+
"learning_rate": 1.6757322175732218e-05,
|
| 588 |
+
"loss": 0.6675,
|
| 589 |
+
"step": 810
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.0186393289841567,
|
| 593 |
+
"grad_norm": 8.320352554321289,
|
| 594 |
+
"learning_rate": 1.6652719665271968e-05,
|
| 595 |
+
"loss": 0.7224,
|
| 596 |
+
"step": 820
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.0310655483069275,
|
| 600 |
+
"grad_norm": 5.392132759094238,
|
| 601 |
+
"learning_rate": 1.6548117154811715e-05,
|
| 602 |
+
"loss": 0.7079,
|
| 603 |
+
"step": 830
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.0434917676296986,
|
| 607 |
+
"grad_norm": 8.554080963134766,
|
| 608 |
+
"learning_rate": 1.6443514644351466e-05,
|
| 609 |
+
"loss": 0.7505,
|
| 610 |
+
"step": 840
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 1.0559179869524697,
|
| 614 |
+
"grad_norm": 9.378899574279785,
|
| 615 |
+
"learning_rate": 1.6338912133891213e-05,
|
| 616 |
+
"loss": 0.6287,
|
| 617 |
+
"step": 850
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.0683442062752408,
|
| 621 |
+
"grad_norm": 7.220132350921631,
|
| 622 |
+
"learning_rate": 1.6234309623430963e-05,
|
| 623 |
+
"loss": 0.6644,
|
| 624 |
+
"step": 860
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.0807704255980117,
|
| 628 |
+
"grad_norm": 8.708951950073242,
|
| 629 |
+
"learning_rate": 1.612970711297071e-05,
|
| 630 |
+
"loss": 0.7393,
|
| 631 |
+
"step": 870
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 1.0931966449207828,
|
| 635 |
+
"grad_norm": 8.052141189575195,
|
| 636 |
+
"learning_rate": 1.602510460251046e-05,
|
| 637 |
+
"loss": 0.7438,
|
| 638 |
+
"step": 880
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.105622864243554,
|
| 642 |
+
"grad_norm": 7.614938735961914,
|
| 643 |
+
"learning_rate": 1.592050209205021e-05,
|
| 644 |
+
"loss": 0.6631,
|
| 645 |
+
"step": 890
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 1.118049083566325,
|
| 649 |
+
"grad_norm": 10.100054740905762,
|
| 650 |
+
"learning_rate": 1.5815899581589958e-05,
|
| 651 |
+
"loss": 0.5594,
|
| 652 |
+
"step": 900
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 1.130475302889096,
|
| 656 |
+
"grad_norm": 8.444486618041992,
|
| 657 |
+
"learning_rate": 1.571129707112971e-05,
|
| 658 |
+
"loss": 0.6603,
|
| 659 |
+
"step": 910
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.142901522211867,
|
| 663 |
+
"grad_norm": 5.679397106170654,
|
| 664 |
+
"learning_rate": 1.560669456066946e-05,
|
| 665 |
+
"loss": 0.6457,
|
| 666 |
+
"step": 920
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 1.1553277415346381,
|
| 670 |
+
"grad_norm": 5.39376163482666,
|
| 671 |
+
"learning_rate": 1.5502092050209206e-05,
|
| 672 |
+
"loss": 0.6546,
|
| 673 |
+
"step": 930
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 1.167753960857409,
|
| 677 |
+
"grad_norm": 8.007966041564941,
|
| 678 |
+
"learning_rate": 1.5397489539748957e-05,
|
| 679 |
+
"loss": 0.7291,
|
| 680 |
+
"step": 940
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.1801801801801801,
|
| 684 |
+
"grad_norm": 8.760208129882812,
|
| 685 |
+
"learning_rate": 1.5292887029288704e-05,
|
| 686 |
+
"loss": 0.688,
|
| 687 |
+
"step": 950
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 1.1926063995029512,
|
| 691 |
+
"grad_norm": 6.147202491760254,
|
| 692 |
+
"learning_rate": 1.5188284518828454e-05,
|
| 693 |
+
"loss": 0.6608,
|
| 694 |
+
"step": 960
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 1.2050326188257223,
|
| 698 |
+
"grad_norm": 8.280518531799316,
|
| 699 |
+
"learning_rate": 1.5083682008368201e-05,
|
| 700 |
+
"loss": 0.6812,
|
| 701 |
+
"step": 970
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 1.2174588381484934,
|
| 705 |
+
"grad_norm": 5.4499053955078125,
|
| 706 |
+
"learning_rate": 1.4979079497907951e-05,
|
| 707 |
+
"loss": 0.6354,
|
| 708 |
+
"step": 980
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 1.2298850574712643,
|
| 712 |
+
"grad_norm": 9.76456356048584,
|
| 713 |
+
"learning_rate": 1.4874476987447699e-05,
|
| 714 |
+
"loss": 0.6708,
|
| 715 |
+
"step": 990
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 1.2423112767940354,
|
| 719 |
+
"grad_norm": 10.97257137298584,
|
| 720 |
+
"learning_rate": 1.4769874476987449e-05,
|
| 721 |
+
"loss": 0.6937,
|
| 722 |
+
"step": 1000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.2547374961168065,
|
| 726 |
+
"grad_norm": 11.25165843963623,
|
| 727 |
+
"learning_rate": 1.4665271966527198e-05,
|
| 728 |
+
"loss": 0.7248,
|
| 729 |
+
"step": 1010
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.2671637154395774,
|
| 733 |
+
"grad_norm": 8.685588836669922,
|
| 734 |
+
"learning_rate": 1.4560669456066946e-05,
|
| 735 |
+
"loss": 0.7222,
|
| 736 |
+
"step": 1020
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.2795899347623485,
|
| 740 |
+
"grad_norm": 7.983170032501221,
|
| 741 |
+
"learning_rate": 1.4456066945606695e-05,
|
| 742 |
+
"loss": 0.7155,
|
| 743 |
+
"step": 1030
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 1.2920161540851196,
|
| 747 |
+
"grad_norm": 3.756377935409546,
|
| 748 |
+
"learning_rate": 1.4351464435146446e-05,
|
| 749 |
+
"loss": 0.6588,
|
| 750 |
+
"step": 1040
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 1.3044423734078907,
|
| 754 |
+
"grad_norm": 5.793786525726318,
|
| 755 |
+
"learning_rate": 1.4246861924686193e-05,
|
| 756 |
+
"loss": 0.6079,
|
| 757 |
+
"step": 1050
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 1.3168685927306618,
|
| 761 |
+
"grad_norm": 6.8630290031433105,
|
| 762 |
+
"learning_rate": 1.4142259414225943e-05,
|
| 763 |
+
"loss": 0.6038,
|
| 764 |
+
"step": 1060
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 1.3292948120534327,
|
| 768 |
+
"grad_norm": 7.02629280090332,
|
| 769 |
+
"learning_rate": 1.403765690376569e-05,
|
| 770 |
+
"loss": 0.6587,
|
| 771 |
+
"step": 1070
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 1.3417210313762038,
|
| 775 |
+
"grad_norm": 5.012918949127197,
|
| 776 |
+
"learning_rate": 1.393305439330544e-05,
|
| 777 |
+
"loss": 0.6479,
|
| 778 |
+
"step": 1080
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 1.354147250698975,
|
| 782 |
+
"grad_norm": 9.064984321594238,
|
| 783 |
+
"learning_rate": 1.382845188284519e-05,
|
| 784 |
+
"loss": 0.6771,
|
| 785 |
+
"step": 1090
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 1.3665734700217458,
|
| 789 |
+
"grad_norm": 7.3286638259887695,
|
| 790 |
+
"learning_rate": 1.3723849372384938e-05,
|
| 791 |
+
"loss": 0.6645,
|
| 792 |
+
"step": 1100
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 1.3789996893445169,
|
| 796 |
+
"grad_norm": 10.485735893249512,
|
| 797 |
+
"learning_rate": 1.3619246861924687e-05,
|
| 798 |
+
"loss": 0.7744,
|
| 799 |
+
"step": 1110
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 1.391425908667288,
|
| 803 |
+
"grad_norm": 8.652807235717773,
|
| 804 |
+
"learning_rate": 1.3514644351464437e-05,
|
| 805 |
+
"loss": 0.6424,
|
| 806 |
+
"step": 1120
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 1.403852127990059,
|
| 810 |
+
"grad_norm": 11.628154754638672,
|
| 811 |
+
"learning_rate": 1.3410041841004184e-05,
|
| 812 |
+
"loss": 0.7039,
|
| 813 |
+
"step": 1130
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 1.4162783473128302,
|
| 817 |
+
"grad_norm": 10.061933517456055,
|
| 818 |
+
"learning_rate": 1.3305439330543935e-05,
|
| 819 |
+
"loss": 0.676,
|
| 820 |
+
"step": 1140
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 1.428704566635601,
|
| 824 |
+
"grad_norm": 5.85321044921875,
|
| 825 |
+
"learning_rate": 1.3200836820083682e-05,
|
| 826 |
+
"loss": 0.6032,
|
| 827 |
+
"step": 1150
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.4411307859583722,
|
| 831 |
+
"grad_norm": 5.4685378074646,
|
| 832 |
+
"learning_rate": 1.3096234309623432e-05,
|
| 833 |
+
"loss": 0.6062,
|
| 834 |
+
"step": 1160
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.4535570052811433,
|
| 838 |
+
"grad_norm": 7.55083703994751,
|
| 839 |
+
"learning_rate": 1.2991631799163181e-05,
|
| 840 |
+
"loss": 0.6231,
|
| 841 |
+
"step": 1170
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.4659832246039142,
|
| 845 |
+
"grad_norm": 9.285325050354004,
|
| 846 |
+
"learning_rate": 1.288702928870293e-05,
|
| 847 |
+
"loss": 0.6233,
|
| 848 |
+
"step": 1180
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.4784094439266853,
|
| 852 |
+
"grad_norm": 8.982355117797852,
|
| 853 |
+
"learning_rate": 1.2782426778242678e-05,
|
| 854 |
+
"loss": 0.5726,
|
| 855 |
+
"step": 1190
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 1.4908356632494564,
|
| 859 |
+
"grad_norm": 17.36650848388672,
|
| 860 |
+
"learning_rate": 1.2677824267782429e-05,
|
| 861 |
+
"loss": 0.5925,
|
| 862 |
+
"step": 1200
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 1.5032618825722275,
|
| 866 |
+
"grad_norm": 6.095853805541992,
|
| 867 |
+
"learning_rate": 1.2573221757322176e-05,
|
| 868 |
+
"loss": 0.6963,
|
| 869 |
+
"step": 1210
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.5156881018949986,
|
| 873 |
+
"grad_norm": 4.788658618927002,
|
| 874 |
+
"learning_rate": 1.2468619246861926e-05,
|
| 875 |
+
"loss": 0.6042,
|
| 876 |
+
"step": 1220
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.5281143212177695,
|
| 880 |
+
"grad_norm": 15.81484317779541,
|
| 881 |
+
"learning_rate": 1.2364016736401673e-05,
|
| 882 |
+
"loss": 0.6201,
|
| 883 |
+
"step": 1230
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.5405405405405406,
|
| 887 |
+
"grad_norm": 8.392333030700684,
|
| 888 |
+
"learning_rate": 1.2259414225941424e-05,
|
| 889 |
+
"loss": 0.718,
|
| 890 |
+
"step": 1240
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.5529667598633115,
|
| 894 |
+
"grad_norm": 9.427897453308105,
|
| 895 |
+
"learning_rate": 1.2154811715481171e-05,
|
| 896 |
+
"loss": 0.6265,
|
| 897 |
+
"step": 1250
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.5653929791860826,
|
| 901 |
+
"grad_norm": 5.241062641143799,
|
| 902 |
+
"learning_rate": 1.2050209205020921e-05,
|
| 903 |
+
"loss": 0.6071,
|
| 904 |
+
"step": 1260
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.5778191985088537,
|
| 908 |
+
"grad_norm": 7.011557579040527,
|
| 909 |
+
"learning_rate": 1.194560669456067e-05,
|
| 910 |
+
"loss": 0.6251,
|
| 911 |
+
"step": 1270
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.5902454178316248,
|
| 915 |
+
"grad_norm": 5.581645965576172,
|
| 916 |
+
"learning_rate": 1.184100418410042e-05,
|
| 917 |
+
"loss": 0.6214,
|
| 918 |
+
"step": 1280
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 1.6026716371543959,
|
| 922 |
+
"grad_norm": 11.603231430053711,
|
| 923 |
+
"learning_rate": 1.1736401673640168e-05,
|
| 924 |
+
"loss": 0.6024,
|
| 925 |
+
"step": 1290
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 1.615097856477167,
|
| 929 |
+
"grad_norm": 9.178550720214844,
|
| 930 |
+
"learning_rate": 1.1631799163179918e-05,
|
| 931 |
+
"loss": 0.6294,
|
| 932 |
+
"step": 1300
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.6275240757999379,
|
| 936 |
+
"grad_norm": 5.397329330444336,
|
| 937 |
+
"learning_rate": 1.1527196652719665e-05,
|
| 938 |
+
"loss": 0.6429,
|
| 939 |
+
"step": 1310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.639950295122709,
|
| 943 |
+
"grad_norm": 6.596381664276123,
|
| 944 |
+
"learning_rate": 1.1422594142259416e-05,
|
| 945 |
+
"loss": 0.6326,
|
| 946 |
+
"step": 1320
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.6523765144454798,
|
| 950 |
+
"grad_norm": 10.257133483886719,
|
| 951 |
+
"learning_rate": 1.1317991631799163e-05,
|
| 952 |
+
"loss": 0.626,
|
| 953 |
+
"step": 1330
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 1.664802733768251,
|
| 957 |
+
"grad_norm": 9.520708084106445,
|
| 958 |
+
"learning_rate": 1.1213389121338913e-05,
|
| 959 |
+
"loss": 0.6784,
|
| 960 |
+
"step": 1340
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 1.677228953091022,
|
| 964 |
+
"grad_norm": 8.193737030029297,
|
| 965 |
+
"learning_rate": 1.1108786610878662e-05,
|
| 966 |
+
"loss": 0.67,
|
| 967 |
+
"step": 1350
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 1.6896551724137931,
|
| 971 |
+
"grad_norm": 10.557798385620117,
|
| 972 |
+
"learning_rate": 1.100418410041841e-05,
|
| 973 |
+
"loss": 0.6265,
|
| 974 |
+
"step": 1360
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.7020813917365643,
|
| 978 |
+
"grad_norm": 5.501838684082031,
|
| 979 |
+
"learning_rate": 1.089958158995816e-05,
|
| 980 |
+
"loss": 0.6588,
|
| 981 |
+
"step": 1370
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 1.7145076110593354,
|
| 985 |
+
"grad_norm": 9.010986328125,
|
| 986 |
+
"learning_rate": 1.079497907949791e-05,
|
| 987 |
+
"loss": 0.6238,
|
| 988 |
+
"step": 1380
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.7269338303821062,
|
| 992 |
+
"grad_norm": 8.921228408813477,
|
| 993 |
+
"learning_rate": 1.0690376569037657e-05,
|
| 994 |
+
"loss": 0.669,
|
| 995 |
+
"step": 1390
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 1.7393600497048773,
|
| 999 |
+
"grad_norm": 8.683089256286621,
|
| 1000 |
+
"learning_rate": 1.0585774058577407e-05,
|
| 1001 |
+
"loss": 0.5655,
|
| 1002 |
+
"step": 1400
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 1.7517862690276482,
|
| 1006 |
+
"grad_norm": 11.0118408203125,
|
| 1007 |
+
"learning_rate": 1.0481171548117154e-05,
|
| 1008 |
+
"loss": 0.6968,
|
| 1009 |
+
"step": 1410
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.7642124883504193,
|
| 1013 |
+
"grad_norm": 7.634212970733643,
|
| 1014 |
+
"learning_rate": 1.0376569037656905e-05,
|
| 1015 |
+
"loss": 0.5995,
|
| 1016 |
+
"step": 1420
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.7766387076731904,
|
| 1020 |
+
"grad_norm": 7.622481346130371,
|
| 1021 |
+
"learning_rate": 1.0271966527196653e-05,
|
| 1022 |
+
"loss": 0.67,
|
| 1023 |
+
"step": 1430
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 1.7890649269959615,
|
| 1027 |
+
"grad_norm": 9.09194278717041,
|
| 1028 |
+
"learning_rate": 1.0167364016736402e-05,
|
| 1029 |
+
"loss": 0.6026,
|
| 1030 |
+
"step": 1440
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 1.8014911463187326,
|
| 1034 |
+
"grad_norm": 8.283378601074219,
|
| 1035 |
+
"learning_rate": 1.0062761506276151e-05,
|
| 1036 |
+
"loss": 0.6119,
|
| 1037 |
+
"step": 1450
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 1.8139173656415035,
|
| 1041 |
+
"grad_norm": 6.903816223144531,
|
| 1042 |
+
"learning_rate": 9.958158995815901e-06,
|
| 1043 |
+
"loss": 0.6326,
|
| 1044 |
+
"step": 1460
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.8263435849642746,
|
| 1048 |
+
"grad_norm": 6.750952243804932,
|
| 1049 |
+
"learning_rate": 9.85355648535565e-06,
|
| 1050 |
+
"loss": 0.6707,
|
| 1051 |
+
"step": 1470
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 1.8387698042870455,
|
| 1055 |
+
"grad_norm": 7.868640899658203,
|
| 1056 |
+
"learning_rate": 9.748953974895399e-06,
|
| 1057 |
+
"loss": 0.6488,
|
| 1058 |
+
"step": 1480
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 1.8511960236098166,
|
| 1062 |
+
"grad_norm": 7.722754001617432,
|
| 1063 |
+
"learning_rate": 9.644351464435148e-06,
|
| 1064 |
+
"loss": 0.6477,
|
| 1065 |
+
"step": 1490
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 1.8636222429325877,
|
| 1069 |
+
"grad_norm": 5.540498733520508,
|
| 1070 |
+
"learning_rate": 9.539748953974896e-06,
|
| 1071 |
+
"loss": 0.6674,
|
| 1072 |
+
"step": 1500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 1.8760484622553588,
|
| 1076 |
+
"grad_norm": 8.326912879943848,
|
| 1077 |
+
"learning_rate": 9.435146443514645e-06,
|
| 1078 |
+
"loss": 0.5612,
|
| 1079 |
+
"step": 1510
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 1.88847468157813,
|
| 1083 |
+
"grad_norm": 10.114307403564453,
|
| 1084 |
+
"learning_rate": 9.330543933054394e-06,
|
| 1085 |
+
"loss": 0.6207,
|
| 1086 |
+
"step": 1520
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 1.900900900900901,
|
| 1090 |
+
"grad_norm": 9.120221138000488,
|
| 1091 |
+
"learning_rate": 9.225941422594144e-06,
|
| 1092 |
+
"loss": 0.5838,
|
| 1093 |
+
"step": 1530
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 1.913327120223672,
|
| 1097 |
+
"grad_norm": 9.336804389953613,
|
| 1098 |
+
"learning_rate": 9.121338912133893e-06,
|
| 1099 |
+
"loss": 0.5486,
|
| 1100 |
+
"step": 1540
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 1.925753339546443,
|
| 1104 |
+
"grad_norm": 6.764523506164551,
|
| 1105 |
+
"learning_rate": 9.016736401673642e-06,
|
| 1106 |
+
"loss": 0.6323,
|
| 1107 |
+
"step": 1550
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 1.938179558869214,
|
| 1111 |
+
"grad_norm": 9.537332534790039,
|
| 1112 |
+
"learning_rate": 8.91213389121339e-06,
|
| 1113 |
+
"loss": 0.5779,
|
| 1114 |
+
"step": 1560
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 1.950605778191985,
|
| 1118 |
+
"grad_norm": 7.732354164123535,
|
| 1119 |
+
"learning_rate": 8.80753138075314e-06,
|
| 1120 |
+
"loss": 0.6373,
|
| 1121 |
+
"step": 1570
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 1.963031997514756,
|
| 1125 |
+
"grad_norm": 10.750661849975586,
|
| 1126 |
+
"learning_rate": 8.702928870292888e-06,
|
| 1127 |
+
"loss": 0.662,
|
| 1128 |
+
"step": 1580
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.9754582168375272,
|
| 1132 |
+
"grad_norm": 5.610159397125244,
|
| 1133 |
+
"learning_rate": 8.598326359832637e-06,
|
| 1134 |
+
"loss": 0.6219,
|
| 1135 |
+
"step": 1590
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 1.9878844361602983,
|
| 1139 |
+
"grad_norm": 5.781522274017334,
|
| 1140 |
+
"learning_rate": 8.493723849372385e-06,
|
| 1141 |
+
"loss": 0.5479,
|
| 1142 |
+
"step": 1600
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.0,
|
| 1146 |
+
"grad_norm": 4.716365814208984,
|
| 1147 |
+
"learning_rate": 8.389121338912136e-06,
|
| 1148 |
+
"loss": 0.5645,
|
| 1149 |
+
"step": 1610
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.0,
|
| 1153 |
+
"eval_accuracy": 0.7378474918465601,
|
| 1154 |
+
"eval_f1": 0.7374887766651477,
|
| 1155 |
+
"eval_loss": 0.6107740998268127,
|
| 1156 |
+
"eval_precision": 0.7408097797843906,
|
| 1157 |
+
"eval_recall": 0.7378474918465601,
|
| 1158 |
+
"eval_runtime": 51.7148,
|
| 1159 |
+
"eval_samples_per_second": 124.51,
|
| 1160 |
+
"eval_steps_per_second": 15.566,
|
| 1161 |
+
"step": 1610
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 2.012426219322771,
|
| 1165 |
+
"grad_norm": 8.042501449584961,
|
| 1166 |
+
"learning_rate": 8.284518828451885e-06,
|
| 1167 |
+
"loss": 0.5036,
|
| 1168 |
+
"step": 1620
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 2.024852438645542,
|
| 1172 |
+
"grad_norm": 6.698486804962158,
|
| 1173 |
+
"learning_rate": 8.179916317991633e-06,
|
| 1174 |
+
"loss": 0.4517,
|
| 1175 |
+
"step": 1630
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 2.0372786579683133,
|
| 1179 |
+
"grad_norm": 8.27925968170166,
|
| 1180 |
+
"learning_rate": 8.075313807531382e-06,
|
| 1181 |
+
"loss": 0.5371,
|
| 1182 |
+
"step": 1640
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 2.049704877291084,
|
| 1186 |
+
"grad_norm": 10.414239883422852,
|
| 1187 |
+
"learning_rate": 7.97071129707113e-06,
|
| 1188 |
+
"loss": 0.5277,
|
| 1189 |
+
"step": 1650
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 2.062131096613855,
|
| 1193 |
+
"grad_norm": 6.001161098480225,
|
| 1194 |
+
"learning_rate": 7.86610878661088e-06,
|
| 1195 |
+
"loss": 0.5582,
|
| 1196 |
+
"step": 1660
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 2.074557315936626,
|
| 1200 |
+
"grad_norm": 7.874392032623291,
|
| 1201 |
+
"learning_rate": 7.761506276150628e-06,
|
| 1202 |
+
"loss": 0.4899,
|
| 1203 |
+
"step": 1670
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 2.0869835352593973,
|
| 1207 |
+
"grad_norm": 4.2222442626953125,
|
| 1208 |
+
"learning_rate": 7.656903765690377e-06,
|
| 1209 |
+
"loss": 0.4986,
|
| 1210 |
+
"step": 1680
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 2.0994097545821684,
|
| 1214 |
+
"grad_norm": 5.905672073364258,
|
| 1215 |
+
"learning_rate": 7.552301255230127e-06,
|
| 1216 |
+
"loss": 0.5021,
|
| 1217 |
+
"step": 1690
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.1118359739049395,
|
| 1221 |
+
"grad_norm": 14.975436210632324,
|
| 1222 |
+
"learning_rate": 7.4476987447698746e-06,
|
| 1223 |
+
"loss": 0.5613,
|
| 1224 |
+
"step": 1700
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 2.1242621932277106,
|
| 1228 |
+
"grad_norm": 12.40459156036377,
|
| 1229 |
+
"learning_rate": 7.343096234309623e-06,
|
| 1230 |
+
"loss": 0.5325,
|
| 1231 |
+
"step": 1710
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 2.1366884125504817,
|
| 1235 |
+
"grad_norm": 5.5273003578186035,
|
| 1236 |
+
"learning_rate": 7.238493723849372e-06,
|
| 1237 |
+
"loss": 0.5685,
|
| 1238 |
+
"step": 1720
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 2.1491146318732524,
|
| 1242 |
+
"grad_norm": 6.3481059074401855,
|
| 1243 |
+
"learning_rate": 7.133891213389122e-06,
|
| 1244 |
+
"loss": 0.5014,
|
| 1245 |
+
"step": 1730
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 2.1615408511960235,
|
| 1249 |
+
"grad_norm": 8.708931922912598,
|
| 1250 |
+
"learning_rate": 7.02928870292887e-06,
|
| 1251 |
+
"loss": 0.5203,
|
| 1252 |
+
"step": 1740
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 2.1739670705187946,
|
| 1256 |
+
"grad_norm": 8.427494049072266,
|
| 1257 |
+
"learning_rate": 6.924686192468619e-06,
|
| 1258 |
+
"loss": 0.4399,
|
| 1259 |
+
"step": 1750
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 2.1863932898415657,
|
| 1263 |
+
"grad_norm": 10.424558639526367,
|
| 1264 |
+
"learning_rate": 6.820083682008368e-06,
|
| 1265 |
+
"loss": 0.4665,
|
| 1266 |
+
"step": 1760
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 2.1988195091643368,
|
| 1270 |
+
"grad_norm": 7.098626613616943,
|
| 1271 |
+
"learning_rate": 6.7154811715481175e-06,
|
| 1272 |
+
"loss": 0.5074,
|
| 1273 |
+
"step": 1770
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 2.211245728487108,
|
| 1277 |
+
"grad_norm": 3.9793460369110107,
|
| 1278 |
+
"learning_rate": 6.610878661087866e-06,
|
| 1279 |
+
"loss": 0.4645,
|
| 1280 |
+
"step": 1780
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 2.223671947809879,
|
| 1284 |
+
"grad_norm": 8.203950881958008,
|
| 1285 |
+
"learning_rate": 6.506276150627615e-06,
|
| 1286 |
+
"loss": 0.4477,
|
| 1287 |
+
"step": 1790
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 2.23609816713265,
|
| 1291 |
+
"grad_norm": 7.613973617553711,
|
| 1292 |
+
"learning_rate": 6.401673640167364e-06,
|
| 1293 |
+
"loss": 0.4876,
|
| 1294 |
+
"step": 1800
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 2.2485243864554207,
|
| 1298 |
+
"grad_norm": 19.467050552368164,
|
| 1299 |
+
"learning_rate": 6.297071129707113e-06,
|
| 1300 |
+
"loss": 0.5133,
|
| 1301 |
+
"step": 1810
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 2.260950605778192,
|
| 1305 |
+
"grad_norm": 8.775492668151855,
|
| 1306 |
+
"learning_rate": 6.192468619246862e-06,
|
| 1307 |
+
"loss": 0.4164,
|
| 1308 |
+
"step": 1820
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 2.273376825100963,
|
| 1312 |
+
"grad_norm": 8.654706001281738,
|
| 1313 |
+
"learning_rate": 6.087866108786611e-06,
|
| 1314 |
+
"loss": 0.4945,
|
| 1315 |
+
"step": 1830
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 2.285803044423734,
|
| 1319 |
+
"grad_norm": 12.804691314697266,
|
| 1320 |
+
"learning_rate": 5.9832635983263595e-06,
|
| 1321 |
+
"loss": 0.5503,
|
| 1322 |
+
"step": 1840
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.298229263746505,
|
| 1326 |
+
"grad_norm": 12.216941833496094,
|
| 1327 |
+
"learning_rate": 5.878661087866109e-06,
|
| 1328 |
+
"loss": 0.3982,
|
| 1329 |
+
"step": 1850
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.3106554830692763,
|
| 1333 |
+
"grad_norm": 18.55878448486328,
|
| 1334 |
+
"learning_rate": 5.774058577405858e-06,
|
| 1335 |
+
"loss": 0.445,
|
| 1336 |
+
"step": 1860
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 2.3230817023920474,
|
| 1340 |
+
"grad_norm": 5.845285892486572,
|
| 1341 |
+
"learning_rate": 5.669456066945607e-06,
|
| 1342 |
+
"loss": 0.5189,
|
| 1343 |
+
"step": 1870
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 2.335507921714818,
|
| 1347 |
+
"grad_norm": 17.797863006591797,
|
| 1348 |
+
"learning_rate": 5.564853556485355e-06,
|
| 1349 |
+
"loss": 0.588,
|
| 1350 |
+
"step": 1880
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 2.347934141037589,
|
| 1354 |
+
"grad_norm": 7.905733108520508,
|
| 1355 |
+
"learning_rate": 5.460251046025105e-06,
|
| 1356 |
+
"loss": 0.452,
|
| 1357 |
+
"step": 1890
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 2.3603603603603602,
|
| 1361 |
+
"grad_norm": 18.246810913085938,
|
| 1362 |
+
"learning_rate": 5.355648535564854e-06,
|
| 1363 |
+
"loss": 0.5252,
|
| 1364 |
+
"step": 1900
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 2.3727865796831313,
|
| 1368 |
+
"grad_norm": 11.707246780395508,
|
| 1369 |
+
"learning_rate": 5.251046025104602e-06,
|
| 1370 |
+
"loss": 0.5335,
|
| 1371 |
+
"step": 1910
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 2.3852127990059024,
|
| 1375 |
+
"grad_norm": 6.850584983825684,
|
| 1376 |
+
"learning_rate": 5.146443514644351e-06,
|
| 1377 |
+
"loss": 0.5001,
|
| 1378 |
+
"step": 1920
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 2.3976390183286735,
|
| 1382 |
+
"grad_norm": 11.419228553771973,
|
| 1383 |
+
"learning_rate": 5.041841004184101e-06,
|
| 1384 |
+
"loss": 0.5328,
|
| 1385 |
+
"step": 1930
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 2.4100652376514446,
|
| 1389 |
+
"grad_norm": 14.90636920928955,
|
| 1390 |
+
"learning_rate": 4.9372384937238495e-06,
|
| 1391 |
+
"loss": 0.485,
|
| 1392 |
+
"step": 1940
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 2.4224914569742158,
|
| 1396 |
+
"grad_norm": 10.567830085754395,
|
| 1397 |
+
"learning_rate": 4.832635983263599e-06,
|
| 1398 |
+
"loss": 0.4796,
|
| 1399 |
+
"step": 1950
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 2.434917676296987,
|
| 1403 |
+
"grad_norm": 9.944981575012207,
|
| 1404 |
+
"learning_rate": 4.728033472803348e-06,
|
| 1405 |
+
"loss": 0.4483,
|
| 1406 |
+
"step": 1960
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.4473438956197575,
|
| 1410 |
+
"grad_norm": 5.981288433074951,
|
| 1411 |
+
"learning_rate": 4.6234309623430966e-06,
|
| 1412 |
+
"loss": 0.5343,
|
| 1413 |
+
"step": 1970
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 2.4597701149425286,
|
| 1417 |
+
"grad_norm": 6.971613883972168,
|
| 1418 |
+
"learning_rate": 4.518828451882845e-06,
|
| 1419 |
+
"loss": 0.4702,
|
| 1420 |
+
"step": 1980
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 2.4721963342652997,
|
| 1424 |
+
"grad_norm": 4.043501377105713,
|
| 1425 |
+
"learning_rate": 4.414225941422595e-06,
|
| 1426 |
+
"loss": 0.4892,
|
| 1427 |
+
"step": 1990
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 2.484622553588071,
|
| 1431 |
+
"grad_norm": 5.520942687988281,
|
| 1432 |
+
"learning_rate": 4.309623430962344e-06,
|
| 1433 |
+
"loss": 0.4353,
|
| 1434 |
+
"step": 2000
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.497048772910842,
|
| 1438 |
+
"grad_norm": 5.04071569442749,
|
| 1439 |
+
"learning_rate": 4.205020920502092e-06,
|
| 1440 |
+
"loss": 0.3699,
|
| 1441 |
+
"step": 2010
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 2.509474992233613,
|
| 1445 |
+
"grad_norm": 16.90415382385254,
|
| 1446 |
+
"learning_rate": 4.100418410041841e-06,
|
| 1447 |
+
"loss": 0.4999,
|
| 1448 |
+
"step": 2020
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 2.5219012115563837,
|
| 1452 |
+
"grad_norm": 13.800042152404785,
|
| 1453 |
+
"learning_rate": 3.995815899581591e-06,
|
| 1454 |
+
"loss": 0.5245,
|
| 1455 |
+
"step": 2030
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 2.534327430879155,
|
| 1459 |
+
"grad_norm": 5.867897987365723,
|
| 1460 |
+
"learning_rate": 3.8912133891213395e-06,
|
| 1461 |
+
"loss": 0.562,
|
| 1462 |
+
"step": 2040
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 2.546753650201926,
|
| 1466 |
+
"grad_norm": 8.719225883483887,
|
| 1467 |
+
"learning_rate": 3.786610878661088e-06,
|
| 1468 |
+
"loss": 0.5359,
|
| 1469 |
+
"step": 2050
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 2.559179869524697,
|
| 1473 |
+
"grad_norm": 7.869408130645752,
|
| 1474 |
+
"learning_rate": 3.6820083682008374e-06,
|
| 1475 |
+
"loss": 0.5104,
|
| 1476 |
+
"step": 2060
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 2.571606088847468,
|
| 1480 |
+
"grad_norm": 7.0496439933776855,
|
| 1481 |
+
"learning_rate": 3.577405857740586e-06,
|
| 1482 |
+
"loss": 0.4387,
|
| 1483 |
+
"step": 2070
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 2.584032308170239,
|
| 1487 |
+
"grad_norm": 7.80281925201416,
|
| 1488 |
+
"learning_rate": 3.4728033472803353e-06,
|
| 1489 |
+
"loss": 0.4326,
|
| 1490 |
+
"step": 2080
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 2.5964585274930103,
|
| 1494 |
+
"grad_norm": 14.786439895629883,
|
| 1495 |
+
"learning_rate": 3.368200836820084e-06,
|
| 1496 |
+
"loss": 0.4535,
|
| 1497 |
+
"step": 2090
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 2.6088847468157814,
|
| 1501 |
+
"grad_norm": 10.45617389678955,
|
| 1502 |
+
"learning_rate": 3.263598326359833e-06,
|
| 1503 |
+
"loss": 0.4656,
|
| 1504 |
+
"step": 2100
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 2.6213109661385525,
|
| 1508 |
+
"grad_norm": 7.8351030349731445,
|
| 1509 |
+
"learning_rate": 3.158995815899582e-06,
|
| 1510 |
+
"loss": 0.4878,
|
| 1511 |
+
"step": 2110
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 2.6337371854613236,
|
| 1515 |
+
"grad_norm": 18.282482147216797,
|
| 1516 |
+
"learning_rate": 3.054393305439331e-06,
|
| 1517 |
+
"loss": 0.4791,
|
| 1518 |
+
"step": 2120
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 2.6461634047840943,
|
| 1522 |
+
"grad_norm": 8.089469909667969,
|
| 1523 |
+
"learning_rate": 2.94979079497908e-06,
|
| 1524 |
+
"loss": 0.4939,
|
| 1525 |
+
"step": 2130
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 2.6585896241068654,
|
| 1529 |
+
"grad_norm": 14.668364524841309,
|
| 1530 |
+
"learning_rate": 2.845188284518829e-06,
|
| 1531 |
+
"loss": 0.5524,
|
| 1532 |
+
"step": 2140
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 2.6710158434296365,
|
| 1536 |
+
"grad_norm": 11.891167640686035,
|
| 1537 |
+
"learning_rate": 2.7405857740585778e-06,
|
| 1538 |
+
"loss": 0.4733,
|
| 1539 |
+
"step": 2150
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 2.6834420627524076,
|
| 1543 |
+
"grad_norm": 7.805383682250977,
|
| 1544 |
+
"learning_rate": 2.635983263598327e-06,
|
| 1545 |
+
"loss": 0.4433,
|
| 1546 |
+
"step": 2160
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 2.6958682820751787,
|
| 1550 |
+
"grad_norm": 8.028349876403809,
|
| 1551 |
+
"learning_rate": 2.5313807531380757e-06,
|
| 1552 |
+
"loss": 0.4669,
|
| 1553 |
+
"step": 2170
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 2.70829450139795,
|
| 1557 |
+
"grad_norm": 11.747848510742188,
|
| 1558 |
+
"learning_rate": 2.4267782426778244e-06,
|
| 1559 |
+
"loss": 0.5418,
|
| 1560 |
+
"step": 2180
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 2.7207207207207205,
|
| 1564 |
+
"grad_norm": 9.765915870666504,
|
| 1565 |
+
"learning_rate": 2.322175732217573e-06,
|
| 1566 |
+
"loss": 0.4536,
|
| 1567 |
+
"step": 2190
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 2.7331469400434916,
|
| 1571 |
+
"grad_norm": 6.603035926818848,
|
| 1572 |
+
"learning_rate": 2.2175732217573223e-06,
|
| 1573 |
+
"loss": 0.4071,
|
| 1574 |
+
"step": 2200
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 2.7455731593662627,
|
| 1578 |
+
"grad_norm": 6.990152835845947,
|
| 1579 |
+
"learning_rate": 2.112970711297071e-06,
|
| 1580 |
+
"loss": 0.4432,
|
| 1581 |
+
"step": 2210
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 2.7579993786890338,
|
| 1585 |
+
"grad_norm": 6.12523078918457,
|
| 1586 |
+
"learning_rate": 2.0083682008368202e-06,
|
| 1587 |
+
"loss": 0.4424,
|
| 1588 |
+
"step": 2220
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 2.770425598011805,
|
| 1592 |
+
"grad_norm": 5.92404842376709,
|
| 1593 |
+
"learning_rate": 1.9037656903765692e-06,
|
| 1594 |
+
"loss": 0.5086,
|
| 1595 |
+
"step": 2230
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 2.782851817334576,
|
| 1599 |
+
"grad_norm": 15.80329704284668,
|
| 1600 |
+
"learning_rate": 1.7991631799163181e-06,
|
| 1601 |
+
"loss": 0.464,
|
| 1602 |
+
"step": 2240
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 2.795278036657347,
|
| 1606 |
+
"grad_norm": 15.308343887329102,
|
| 1607 |
+
"learning_rate": 1.694560669456067e-06,
|
| 1608 |
+
"loss": 0.5256,
|
| 1609 |
+
"step": 2250
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 2.807704255980118,
|
| 1613 |
+
"grad_norm": 9.566842079162598,
|
| 1614 |
+
"learning_rate": 1.589958158995816e-06,
|
| 1615 |
+
"loss": 0.477,
|
| 1616 |
+
"step": 2260
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 2.8201304753028893,
|
| 1620 |
+
"grad_norm": 8.725525856018066,
|
| 1621 |
+
"learning_rate": 1.485355648535565e-06,
|
| 1622 |
+
"loss": 0.4643,
|
| 1623 |
+
"step": 2270
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 2.8325566946256604,
|
| 1627 |
+
"grad_norm": 7.611164569854736,
|
| 1628 |
+
"learning_rate": 1.380753138075314e-06,
|
| 1629 |
+
"loss": 0.5102,
|
| 1630 |
+
"step": 2280
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 2.844982913948431,
|
| 1634 |
+
"grad_norm": 7.241260528564453,
|
| 1635 |
+
"learning_rate": 1.276150627615063e-06,
|
| 1636 |
+
"loss": 0.4782,
|
| 1637 |
+
"step": 2290
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 2.857409133271202,
|
| 1641 |
+
"grad_norm": 12.204206466674805,
|
| 1642 |
+
"learning_rate": 1.1715481171548119e-06,
|
| 1643 |
+
"loss": 0.5092,
|
| 1644 |
+
"step": 2300
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 2.8698353525939733,
|
| 1648 |
+
"grad_norm": 11.016090393066406,
|
| 1649 |
+
"learning_rate": 1.0669456066945608e-06,
|
| 1650 |
+
"loss": 0.474,
|
| 1651 |
+
"step": 2310
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 2.8822615719167444,
|
| 1655 |
+
"grad_norm": 6.266747951507568,
|
| 1656 |
+
"learning_rate": 9.623430962343098e-07,
|
| 1657 |
+
"loss": 0.4315,
|
| 1658 |
+
"step": 2320
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 2.8946877912395155,
|
| 1662 |
+
"grad_norm": 11.70011043548584,
|
| 1663 |
+
"learning_rate": 8.577405857740586e-07,
|
| 1664 |
+
"loss": 0.4458,
|
| 1665 |
+
"step": 2330
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 2.9071140105622866,
|
| 1669 |
+
"grad_norm": 8.788187026977539,
|
| 1670 |
+
"learning_rate": 7.531380753138076e-07,
|
| 1671 |
+
"loss": 0.5513,
|
| 1672 |
+
"step": 2340
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 2.9195402298850572,
|
| 1676 |
+
"grad_norm": 7.601880073547363,
|
| 1677 |
+
"learning_rate": 6.485355648535565e-07,
|
| 1678 |
+
"loss": 0.4752,
|
| 1679 |
+
"step": 2350
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 2.9319664492078283,
|
| 1683 |
+
"grad_norm": 9.370701789855957,
|
| 1684 |
+
"learning_rate": 5.439330543933055e-07,
|
| 1685 |
+
"loss": 0.4474,
|
| 1686 |
+
"step": 2360
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 2.9443926685305994,
|
| 1690 |
+
"grad_norm": 8.328364372253418,
|
| 1691 |
+
"learning_rate": 4.393305439330544e-07,
|
| 1692 |
+
"loss": 0.4432,
|
| 1693 |
+
"step": 2370
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 2.9568188878533705,
|
| 1697 |
+
"grad_norm": 8.796112060546875,
|
| 1698 |
+
"learning_rate": 3.3472803347280335e-07,
|
| 1699 |
+
"loss": 0.4615,
|
| 1700 |
+
"step": 2380
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 2.9692451071761417,
|
| 1704 |
+
"grad_norm": 7.252564907073975,
|
| 1705 |
+
"learning_rate": 2.3012552301255234e-07,
|
| 1706 |
+
"loss": 0.5135,
|
| 1707 |
+
"step": 2390
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 2.9816713264989128,
|
| 1711 |
+
"grad_norm": 7.336194038391113,
|
| 1712 |
+
"learning_rate": 1.2552301255230126e-07,
|
| 1713 |
+
"loss": 0.4412,
|
| 1714 |
+
"step": 2400
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 2.994097545821684,
|
| 1718 |
+
"grad_norm": 7.167020320892334,
|
| 1719 |
+
"learning_rate": 2.092050209205021e-08,
|
| 1720 |
+
"loss": 0.4858,
|
| 1721 |
+
"step": 2410
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 2.996582789686238,
|
| 1725 |
+
"eval_accuracy": 0.7682870010871253,
|
| 1726 |
+
"eval_f1": 0.766151133896168,
|
| 1727 |
+
"eval_loss": 0.5731337070465088,
|
| 1728 |
+
"eval_precision": 0.7674176393114566,
|
| 1729 |
+
"eval_recall": 0.7682870010871253,
|
| 1730 |
+
"eval_runtime": 52.1315,
|
| 1731 |
+
"eval_samples_per_second": 123.515,
|
| 1732 |
+
"eval_steps_per_second": 15.442,
|
| 1733 |
+
"step": 2412
|
| 1734 |
+
}
|
| 1735 |
+
],
|
| 1736 |
+
"logging_steps": 10,
|
| 1737 |
+
"max_steps": 2412,
|
| 1738 |
+
"num_input_tokens_seen": 0,
|
| 1739 |
+
"num_train_epochs": 3,
|
| 1740 |
+
"save_steps": 500,
|
| 1741 |
+
"stateful_callbacks": {
|
| 1742 |
+
"EarlyStoppingCallback": {
|
| 1743 |
+
"args": {
|
| 1744 |
+
"early_stopping_patience": 3,
|
| 1745 |
+
"early_stopping_threshold": 0.0
|
| 1746 |
+
},
|
| 1747 |
+
"attributes": {
|
| 1748 |
+
"early_stopping_patience_counter": 0
|
| 1749 |
+
}
|
| 1750 |
+
},
|
| 1751 |
+
"TrainerControl": {
|
| 1752 |
+
"args": {
|
| 1753 |
+
"should_epoch_stop": false,
|
| 1754 |
+
"should_evaluate": false,
|
| 1755 |
+
"should_log": false,
|
| 1756 |
+
"should_save": true,
|
| 1757 |
+
"should_training_stop": true
|
| 1758 |
+
},
|
| 1759 |
+
"attributes": {}
|
| 1760 |
+
}
|
| 1761 |
+
},
|
| 1762 |
+
"total_flos": 555185755994688.0,
|
| 1763 |
+
"train_batch_size": 8,
|
| 1764 |
+
"trial_name": null,
|
| 1765 |
+
"trial_params": null
|
| 1766 |
+
}
|
checkpoint-2412/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b74a6de7f3ed0719b44759cf4f2718a423a057052a06b70b41090640f4b42b14
|
| 3 |
+
size 5304
|
checkpoint-805/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"directionality": "bidi",
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"LABEL_0": 0,
|
| 20 |
+
"LABEL_1": 1,
|
| 21 |
+
"LABEL_2": 2
|
| 22 |
+
},
|
| 23 |
+
"layer_norm_eps": 1e-12,
|
| 24 |
+
"max_position_embeddings": 512,
|
| 25 |
+
"model_type": "bert",
|
| 26 |
+
"num_attention_heads": 12,
|
| 27 |
+
"num_hidden_layers": 12,
|
| 28 |
+
"output_past": true,
|
| 29 |
+
"pad_token_id": 0,
|
| 30 |
+
"pooler_fc_size": 768,
|
| 31 |
+
"pooler_num_attention_heads": 12,
|
| 32 |
+
"pooler_num_fc_layers": 3,
|
| 33 |
+
"pooler_size_per_head": 128,
|
| 34 |
+
"pooler_type": "first_token_transform",
|
| 35 |
+
"position_embedding_type": "absolute",
|
| 36 |
+
"problem_type": "single_label_classification",
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.50.2",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 29794
|
| 42 |
+
}
|
checkpoint-805/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd7a51897a4e15e57a76ac5894a496576bf7244bdeb32c827a1edc35c8e0b39c
|
| 3 |
+
size 435725300
|
checkpoint-805/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8304fe88f34e7febbb0ee39daa9b5186e83ecfcee07bbcb1cae8ee5272169f17
|
| 3 |
+
size 871566074
|
checkpoint-805/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6041670accee4b0beaa30b42e20f42524de8577748699f5b84762e616f853b1
|
| 3 |
+
size 13990
|
checkpoint-805/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:50937ea91d8bc0d142fd3eda2dc23ba5584028ffb3340deee6467d1b70e256e8
|
| 3 |
+
size 1064
|
checkpoint-805/trainer_state.json
ADDED
|
@@ -0,0 +1,615 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 805,
|
| 3 |
+
"best_metric": 0.6694435151334517,
|
| 4 |
+
"best_model_checkpoint": "./results\\checkpoint-805",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 805,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.012426219322771047,
|
| 14 |
+
"grad_norm": 7.803389072418213,
|
| 15 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 16 |
+
"loss": 1.0963,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.024852438645542093,
|
| 21 |
+
"grad_norm": 3.206519842147827,
|
| 22 |
+
"learning_rate": 8.000000000000001e-07,
|
| 23 |
+
"loss": 1.0925,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03727865796831314,
|
| 28 |
+
"grad_norm": 6.171759605407715,
|
| 29 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 30 |
+
"loss": 1.082,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.049704877291084186,
|
| 35 |
+
"grad_norm": 5.720030784606934,
|
| 36 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 37 |
+
"loss": 1.0757,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.062131096613855234,
|
| 42 |
+
"grad_norm": 2.8826277256011963,
|
| 43 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 44 |
+
"loss": 1.0982,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.07455731593662628,
|
| 49 |
+
"grad_norm": 3.8596413135528564,
|
| 50 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 51 |
+
"loss": 1.0692,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.08698353525939732,
|
| 56 |
+
"grad_norm": 1.860378623008728,
|
| 57 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 58 |
+
"loss": 1.0862,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.09940975458216837,
|
| 63 |
+
"grad_norm": 1.8722461462020874,
|
| 64 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 65 |
+
"loss": 1.0569,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.11183597390493942,
|
| 70 |
+
"grad_norm": 3.263986349105835,
|
| 71 |
+
"learning_rate": 3.6000000000000003e-06,
|
| 72 |
+
"loss": 1.0725,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.12426219322771047,
|
| 77 |
+
"grad_norm": 2.589475393295288,
|
| 78 |
+
"learning_rate": 4.000000000000001e-06,
|
| 79 |
+
"loss": 1.0447,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1366884125504815,
|
| 84 |
+
"grad_norm": 1.8939751386642456,
|
| 85 |
+
"learning_rate": 4.4e-06,
|
| 86 |
+
"loss": 1.056,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.14911463187325255,
|
| 91 |
+
"grad_norm": 2.082205057144165,
|
| 92 |
+
"learning_rate": 4.800000000000001e-06,
|
| 93 |
+
"loss": 1.0228,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1615408511960236,
|
| 98 |
+
"grad_norm": 3.9843056201934814,
|
| 99 |
+
"learning_rate": 5.2e-06,
|
| 100 |
+
"loss": 1.021,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.17396707051879465,
|
| 105 |
+
"grad_norm": 3.245530128479004,
|
| 106 |
+
"learning_rate": 5.600000000000001e-06,
|
| 107 |
+
"loss": 1.0255,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1863932898415657,
|
| 112 |
+
"grad_norm": 3.017813205718994,
|
| 113 |
+
"learning_rate": 6e-06,
|
| 114 |
+
"loss": 1.016,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.19881950916433674,
|
| 119 |
+
"grad_norm": 3.6596145629882812,
|
| 120 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 121 |
+
"loss": 0.9506,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.2112457284871078,
|
| 126 |
+
"grad_norm": 3.5496115684509277,
|
| 127 |
+
"learning_rate": 6.800000000000001e-06,
|
| 128 |
+
"loss": 0.9531,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.22367194780987884,
|
| 133 |
+
"grad_norm": 7.028264045715332,
|
| 134 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 135 |
+
"loss": 0.9606,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.2360981671326499,
|
| 140 |
+
"grad_norm": 4.832308292388916,
|
| 141 |
+
"learning_rate": 7.600000000000001e-06,
|
| 142 |
+
"loss": 0.9493,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.24852438645542094,
|
| 147 |
+
"grad_norm": 5.981413841247559,
|
| 148 |
+
"learning_rate": 8.000000000000001e-06,
|
| 149 |
+
"loss": 0.9692,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.26095060577819196,
|
| 154 |
+
"grad_norm": 4.8830952644348145,
|
| 155 |
+
"learning_rate": 8.400000000000001e-06,
|
| 156 |
+
"loss": 0.8339,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.273376825100963,
|
| 161 |
+
"grad_norm": 7.470656394958496,
|
| 162 |
+
"learning_rate": 8.8e-06,
|
| 163 |
+
"loss": 0.8902,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.28580304442373405,
|
| 168 |
+
"grad_norm": 10.00381088256836,
|
| 169 |
+
"learning_rate": 9.200000000000002e-06,
|
| 170 |
+
"loss": 0.9514,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2982292637465051,
|
| 175 |
+
"grad_norm": 8.733065605163574,
|
| 176 |
+
"learning_rate": 9.600000000000001e-06,
|
| 177 |
+
"loss": 0.9589,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.31065548306927615,
|
| 182 |
+
"grad_norm": 9.063344955444336,
|
| 183 |
+
"learning_rate": 1e-05,
|
| 184 |
+
"loss": 0.9367,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.3230817023920472,
|
| 189 |
+
"grad_norm": 4.745977401733398,
|
| 190 |
+
"learning_rate": 1.04e-05,
|
| 191 |
+
"loss": 0.8591,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.33550792171481825,
|
| 196 |
+
"grad_norm": 6.759552001953125,
|
| 197 |
+
"learning_rate": 1.0800000000000002e-05,
|
| 198 |
+
"loss": 0.8669,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.3479341410375893,
|
| 203 |
+
"grad_norm": 6.986461639404297,
|
| 204 |
+
"learning_rate": 1.1200000000000001e-05,
|
| 205 |
+
"loss": 0.8973,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.36036036036036034,
|
| 210 |
+
"grad_norm": 9.381807327270508,
|
| 211 |
+
"learning_rate": 1.16e-05,
|
| 212 |
+
"loss": 0.8768,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.3727865796831314,
|
| 217 |
+
"grad_norm": 9.528881072998047,
|
| 218 |
+
"learning_rate": 1.2e-05,
|
| 219 |
+
"loss": 0.9337,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.38521279900590244,
|
| 224 |
+
"grad_norm": 5.089269638061523,
|
| 225 |
+
"learning_rate": 1.2400000000000002e-05,
|
| 226 |
+
"loss": 0.8917,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3976390183286735,
|
| 231 |
+
"grad_norm": 5.263373374938965,
|
| 232 |
+
"learning_rate": 1.2800000000000001e-05,
|
| 233 |
+
"loss": 0.8682,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.41006523765144454,
|
| 238 |
+
"grad_norm": 5.064725399017334,
|
| 239 |
+
"learning_rate": 1.3200000000000002e-05,
|
| 240 |
+
"loss": 0.9005,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.4224914569742156,
|
| 245 |
+
"grad_norm": 7.281830787658691,
|
| 246 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 247 |
+
"loss": 0.83,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.43491767629698663,
|
| 252 |
+
"grad_norm": 6.057513236999512,
|
| 253 |
+
"learning_rate": 1.4e-05,
|
| 254 |
+
"loss": 0.8947,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.4473438956197577,
|
| 259 |
+
"grad_norm": 7.247957229614258,
|
| 260 |
+
"learning_rate": 1.4400000000000001e-05,
|
| 261 |
+
"loss": 0.9045,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.45977011494252873,
|
| 266 |
+
"grad_norm": 5.190288066864014,
|
| 267 |
+
"learning_rate": 1.48e-05,
|
| 268 |
+
"loss": 0.8841,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4721963342652998,
|
| 273 |
+
"grad_norm": 7.859437465667725,
|
| 274 |
+
"learning_rate": 1.5200000000000002e-05,
|
| 275 |
+
"loss": 0.8824,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.4846225535880708,
|
| 280 |
+
"grad_norm": 4.713529586791992,
|
| 281 |
+
"learning_rate": 1.5600000000000003e-05,
|
| 282 |
+
"loss": 0.8779,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.4970487729108419,
|
| 287 |
+
"grad_norm": 5.651578903198242,
|
| 288 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 289 |
+
"loss": 0.868,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.5094749922336129,
|
| 294 |
+
"grad_norm": 8.400832176208496,
|
| 295 |
+
"learning_rate": 1.64e-05,
|
| 296 |
+
"loss": 0.86,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.5219012115563839,
|
| 301 |
+
"grad_norm": 4.4787068367004395,
|
| 302 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 303 |
+
"loss": 0.865,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.534327430879155,
|
| 308 |
+
"grad_norm": 8.719943046569824,
|
| 309 |
+
"learning_rate": 1.72e-05,
|
| 310 |
+
"loss": 0.848,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.546753650201926,
|
| 315 |
+
"grad_norm": 5.370767593383789,
|
| 316 |
+
"learning_rate": 1.76e-05,
|
| 317 |
+
"loss": 0.8235,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.5591798695246971,
|
| 322 |
+
"grad_norm": 7.650132179260254,
|
| 323 |
+
"learning_rate": 1.8e-05,
|
| 324 |
+
"loss": 0.8323,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5716060888474681,
|
| 329 |
+
"grad_norm": 5.548404693603516,
|
| 330 |
+
"learning_rate": 1.8400000000000003e-05,
|
| 331 |
+
"loss": 0.8242,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5840323081702392,
|
| 336 |
+
"grad_norm": 5.651975631713867,
|
| 337 |
+
"learning_rate": 1.88e-05,
|
| 338 |
+
"loss": 0.7487,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5964585274930102,
|
| 343 |
+
"grad_norm": 4.232389450073242,
|
| 344 |
+
"learning_rate": 1.9200000000000003e-05,
|
| 345 |
+
"loss": 0.8622,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.6088847468157813,
|
| 350 |
+
"grad_norm": 5.476607322692871,
|
| 351 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 352 |
+
"loss": 0.8144,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.6213109661385523,
|
| 357 |
+
"grad_norm": 6.335455894470215,
|
| 358 |
+
"learning_rate": 2e-05,
|
| 359 |
+
"loss": 0.8288,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.6337371854613234,
|
| 364 |
+
"grad_norm": 7.2705912590026855,
|
| 365 |
+
"learning_rate": 1.989539748953975e-05,
|
| 366 |
+
"loss": 0.8573,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6461634047840944,
|
| 371 |
+
"grad_norm": 6.81685209274292,
|
| 372 |
+
"learning_rate": 1.97907949790795e-05,
|
| 373 |
+
"loss": 0.8903,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6585896241068655,
|
| 378 |
+
"grad_norm": 5.642577648162842,
|
| 379 |
+
"learning_rate": 1.9686192468619246e-05,
|
| 380 |
+
"loss": 0.8102,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.6710158434296365,
|
| 385 |
+
"grad_norm": 11.23028564453125,
|
| 386 |
+
"learning_rate": 1.9581589958158997e-05,
|
| 387 |
+
"loss": 0.8068,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.6834420627524076,
|
| 392 |
+
"grad_norm": 10.290377616882324,
|
| 393 |
+
"learning_rate": 1.9476987447698744e-05,
|
| 394 |
+
"loss": 0.8092,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.6958682820751786,
|
| 399 |
+
"grad_norm": 7.7588629722595215,
|
| 400 |
+
"learning_rate": 1.9372384937238494e-05,
|
| 401 |
+
"loss": 0.7837,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.7082945013979497,
|
| 406 |
+
"grad_norm": 6.096302509307861,
|
| 407 |
+
"learning_rate": 1.9267782426778245e-05,
|
| 408 |
+
"loss": 0.8138,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.7207207207207207,
|
| 413 |
+
"grad_norm": 9.541234970092773,
|
| 414 |
+
"learning_rate": 1.916317991631799e-05,
|
| 415 |
+
"loss": 0.8009,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.7331469400434918,
|
| 420 |
+
"grad_norm": 6.361993789672852,
|
| 421 |
+
"learning_rate": 1.9058577405857742e-05,
|
| 422 |
+
"loss": 0.8185,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.7455731593662628,
|
| 427 |
+
"grad_norm": 12.062479972839355,
|
| 428 |
+
"learning_rate": 1.8953974895397492e-05,
|
| 429 |
+
"loss": 0.847,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.7579993786890339,
|
| 434 |
+
"grad_norm": 5.53268575668335,
|
| 435 |
+
"learning_rate": 1.884937238493724e-05,
|
| 436 |
+
"loss": 0.7892,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.7704255980118049,
|
| 441 |
+
"grad_norm": 7.868827819824219,
|
| 442 |
+
"learning_rate": 1.874476987447699e-05,
|
| 443 |
+
"loss": 0.7591,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.782851817334576,
|
| 448 |
+
"grad_norm": 7.4785380363464355,
|
| 449 |
+
"learning_rate": 1.8640167364016737e-05,
|
| 450 |
+
"loss": 0.7875,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.795278036657347,
|
| 455 |
+
"grad_norm": 6.592331409454346,
|
| 456 |
+
"learning_rate": 1.8535564853556487e-05,
|
| 457 |
+
"loss": 0.8145,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.8077042559801181,
|
| 462 |
+
"grad_norm": 6.539604663848877,
|
| 463 |
+
"learning_rate": 1.8430962343096234e-05,
|
| 464 |
+
"loss": 0.7509,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.8201304753028891,
|
| 469 |
+
"grad_norm": 8.498805046081543,
|
| 470 |
+
"learning_rate": 1.8326359832635985e-05,
|
| 471 |
+
"loss": 0.8198,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.8325566946256602,
|
| 476 |
+
"grad_norm": 7.241284370422363,
|
| 477 |
+
"learning_rate": 1.8221757322175732e-05,
|
| 478 |
+
"loss": 0.7804,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.8449829139484312,
|
| 483 |
+
"grad_norm": 8.760014533996582,
|
| 484 |
+
"learning_rate": 1.8117154811715482e-05,
|
| 485 |
+
"loss": 0.8414,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.8574091332712023,
|
| 490 |
+
"grad_norm": 4.911290168762207,
|
| 491 |
+
"learning_rate": 1.801255230125523e-05,
|
| 492 |
+
"loss": 0.788,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.8698353525939733,
|
| 497 |
+
"grad_norm": 5.76509952545166,
|
| 498 |
+
"learning_rate": 1.790794979079498e-05,
|
| 499 |
+
"loss": 0.7133,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.8822615719167444,
|
| 504 |
+
"grad_norm": 6.4433088302612305,
|
| 505 |
+
"learning_rate": 1.7803347280334727e-05,
|
| 506 |
+
"loss": 0.7887,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.8946877912395154,
|
| 511 |
+
"grad_norm": 6.976423740386963,
|
| 512 |
+
"learning_rate": 1.7698744769874477e-05,
|
| 513 |
+
"loss": 0.8825,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.9071140105622865,
|
| 518 |
+
"grad_norm": 5.915661811828613,
|
| 519 |
+
"learning_rate": 1.7594142259414228e-05,
|
| 520 |
+
"loss": 0.727,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.9195402298850575,
|
| 525 |
+
"grad_norm": 6.439208030700684,
|
| 526 |
+
"learning_rate": 1.7489539748953975e-05,
|
| 527 |
+
"loss": 0.7372,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.9319664492078286,
|
| 532 |
+
"grad_norm": 7.3419084548950195,
|
| 533 |
+
"learning_rate": 1.7384937238493725e-05,
|
| 534 |
+
"loss": 0.7954,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.9443926685305996,
|
| 539 |
+
"grad_norm": 6.151190280914307,
|
| 540 |
+
"learning_rate": 1.7280334728033476e-05,
|
| 541 |
+
"loss": 0.7261,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.9568188878533707,
|
| 546 |
+
"grad_norm": 6.01263952255249,
|
| 547 |
+
"learning_rate": 1.7175732217573223e-05,
|
| 548 |
+
"loss": 0.773,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.9692451071761417,
|
| 553 |
+
"grad_norm": 6.289830684661865,
|
| 554 |
+
"learning_rate": 1.7071129707112973e-05,
|
| 555 |
+
"loss": 0.8026,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.9816713264989128,
|
| 560 |
+
"grad_norm": 8.667413711547852,
|
| 561 |
+
"learning_rate": 1.696652719665272e-05,
|
| 562 |
+
"loss": 0.7511,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.9940975458216837,
|
| 567 |
+
"grad_norm": 26.640560150146484,
|
| 568 |
+
"learning_rate": 1.686192468619247e-05,
|
| 569 |
+
"loss": 0.7553,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.0,
|
| 574 |
+
"eval_accuracy": 0.6735517937567945,
|
| 575 |
+
"eval_f1": 0.6694435151334517,
|
| 576 |
+
"eval_loss": 0.7382379770278931,
|
| 577 |
+
"eval_precision": 0.6693546214892159,
|
| 578 |
+
"eval_recall": 0.6735517937567945,
|
| 579 |
+
"eval_runtime": 51.9119,
|
| 580 |
+
"eval_samples_per_second": 124.037,
|
| 581 |
+
"eval_steps_per_second": 15.507,
|
| 582 |
+
"step": 805
|
| 583 |
+
}
|
| 584 |
+
],
|
| 585 |
+
"logging_steps": 10,
|
| 586 |
+
"max_steps": 2412,
|
| 587 |
+
"num_input_tokens_seen": 0,
|
| 588 |
+
"num_train_epochs": 3,
|
| 589 |
+
"save_steps": 500,
|
| 590 |
+
"stateful_callbacks": {
|
| 591 |
+
"EarlyStoppingCallback": {
|
| 592 |
+
"args": {
|
| 593 |
+
"early_stopping_patience": 3,
|
| 594 |
+
"early_stopping_threshold": 0.0
|
| 595 |
+
},
|
| 596 |
+
"attributes": {
|
| 597 |
+
"early_stopping_patience_counter": 0
|
| 598 |
+
}
|
| 599 |
+
},
|
| 600 |
+
"TrainerControl": {
|
| 601 |
+
"args": {
|
| 602 |
+
"should_epoch_stop": false,
|
| 603 |
+
"should_evaluate": false,
|
| 604 |
+
"should_log": false,
|
| 605 |
+
"should_save": true,
|
| 606 |
+
"should_training_stop": false
|
| 607 |
+
},
|
| 608 |
+
"attributes": {}
|
| 609 |
+
}
|
| 610 |
+
},
|
| 611 |
+
"total_flos": 185272957552032.0,
|
| 612 |
+
"train_batch_size": 8,
|
| 613 |
+
"trial_name": null,
|
| 614 |
+
"trial_params": null
|
| 615 |
+
}
|
checkpoint-805/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b74a6de7f3ed0719b44759cf4f2718a423a057052a06b70b41090640f4b42b14
|
| 3 |
+
size 5304
|
config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"directionality": "bidi",
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"LABEL_0": 0,
|
| 20 |
+
"LABEL_1": 1,
|
| 21 |
+
"LABEL_2": 2
|
| 22 |
+
},
|
| 23 |
+
"layer_norm_eps": 1e-12,
|
| 24 |
+
"max_position_embeddings": 512,
|
| 25 |
+
"model_type": "bert",
|
| 26 |
+
"num_attention_heads": 12,
|
| 27 |
+
"num_hidden_layers": 12,
|
| 28 |
+
"output_past": true,
|
| 29 |
+
"pad_token_id": 0,
|
| 30 |
+
"pooler_fc_size": 768,
|
| 31 |
+
"pooler_num_attention_heads": 12,
|
| 32 |
+
"pooler_num_fc_layers": 3,
|
| 33 |
+
"pooler_size_per_head": 128,
|
| 34 |
+
"pooler_type": "first_token_transform",
|
| 35 |
+
"position_embedding_type": "absolute",
|
| 36 |
+
"problem_type": "single_label_classification",
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.50.2",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 29794
|
| 42 |
+
}
|
logs/events.out.tfevents.1743420092.DESKTOP-JBF3D6L.2192.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7bb2a9e88ab87e7c18260ac7b8fd5cf98dcf7ef54f95253f4377971d4ce2027
|
| 3 |
+
size 5323
|
logs/events.out.tfevents.1743420252.DESKTOP-JBF3D6L.6468.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2bfa34bbcb5725ce7cdb222f9bd611efb4de3d2b8b0d547ba2a8bc353b845101
|
| 3 |
+
size 5323
|
logs/events.out.tfevents.1743420302.DESKTOP-JBF3D6L.11468.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1780056e89a40cb9c7b5b4f06edf0b1575319aa6aa8aa1c4e37f6ec51d7aa401
|
| 3 |
+
size 5373
|
logs/events.out.tfevents.1743420355.DESKTOP-JBF3D6L.6752.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad6431404868a23e94f3f888a1ae20fe2657245c800e2c1fa5b3ed9ff4ab7a49
|
| 3 |
+
size 5387
|
logs/events.out.tfevents.1743420703.DESKTOP-JBF3D6L.4224.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d2519c7a6d741b075607b9f3ea2af903388117916413d6edbdc495aeeafa761
|
| 3 |
+
size 5387
|
logs/events.out.tfevents.1743423112.DESKTOP-JBF3D6L.9152.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c4b8c6002e12bf5a70e67f5f199aa625e8fb789aa968ad7a8739fe2a2eb1a99
|
| 3 |
+
size 57960
|
logs/events.out.tfevents.1743427318.DESKTOP-JBF3D6L.9152.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b4af71bcd46852cc805591ba973b83c76cc6877107574007298a1092215669d9
|
| 3 |
+
size 560
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5e2549dabd16a904f8d8e2a6da81bbfd7d4245fca26114dee7a649068123d9f
|
| 3 |
+
size 435725300
|
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_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b74a6de7f3ed0719b44759cf4f2718a423a057052a06b70b41090640f4b42b14
|
| 3 |
+
size 5304
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|