Upload 7 files
Browse files- config.json +25 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1335 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilbert-base-uncased",
|
| 3 |
+
"activation": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"DistilBertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"hidden_dim": 3072,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"max_position_embeddings": 512,
|
| 13 |
+
"model_type": "distilbert",
|
| 14 |
+
"n_heads": 12,
|
| 15 |
+
"n_layers": 6,
|
| 16 |
+
"pad_token_id": 0,
|
| 17 |
+
"problem_type": "single_label_classification",
|
| 18 |
+
"qa_dropout": 0.1,
|
| 19 |
+
"seq_classif_dropout": 0.2,
|
| 20 |
+
"sinusoidal_pos_embds": false,
|
| 21 |
+
"tie_weights_": true,
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.42.4",
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:efa06266c47bc101f1762f52e259bd7967dba4c8960b9c3969b573e89a3ada9a
|
| 3 |
+
size 267832560
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8db22b75a53fdd336fdf00732069a4c4664ebc2c867da031b21f03bcd8894ac
|
| 3 |
+
size 535727290
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b1079af047045f0b7a6ce6f9857f24163cbace39e0b4cbd199dd880c9ea6000
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53135894f6a9a1ab9ae0094e690858a4c0fc3d644ff38737b2d4fbd03b4efd7e
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1335 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1869,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.016051364365971106,
|
| 13 |
+
"grad_norm": 0.5444720983505249,
|
| 14 |
+
"learning_rate": 4.973247726056715e-05,
|
| 15 |
+
"loss": 0.6974,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03210272873194221,
|
| 20 |
+
"grad_norm": 0.46184447407722473,
|
| 21 |
+
"learning_rate": 4.94649545211343e-05,
|
| 22 |
+
"loss": 0.6938,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.048154093097913325,
|
| 27 |
+
"grad_norm": 0.6587526798248291,
|
| 28 |
+
"learning_rate": 4.919743178170145e-05,
|
| 29 |
+
"loss": 0.6926,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06420545746388442,
|
| 34 |
+
"grad_norm": 2.1942267417907715,
|
| 35 |
+
"learning_rate": 4.8929909042268596e-05,
|
| 36 |
+
"loss": 0.6628,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08025682182985554,
|
| 41 |
+
"grad_norm": 1.835483431816101,
|
| 42 |
+
"learning_rate": 4.8662386302835744e-05,
|
| 43 |
+
"loss": 0.6777,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.09630818619582665,
|
| 48 |
+
"grad_norm": 1.6423420906066895,
|
| 49 |
+
"learning_rate": 4.839486356340289e-05,
|
| 50 |
+
"loss": 0.6748,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.11235955056179775,
|
| 55 |
+
"grad_norm": 1.618177056312561,
|
| 56 |
+
"learning_rate": 4.812734082397004e-05,
|
| 57 |
+
"loss": 0.6585,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.12841091492776885,
|
| 62 |
+
"grad_norm": 2.7368061542510986,
|
| 63 |
+
"learning_rate": 4.785981808453719e-05,
|
| 64 |
+
"loss": 0.6208,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.14446227929373998,
|
| 69 |
+
"grad_norm": 2.0427486896514893,
|
| 70 |
+
"learning_rate": 4.759229534510434e-05,
|
| 71 |
+
"loss": 0.6447,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.16051364365971107,
|
| 76 |
+
"grad_norm": 2.044725179672241,
|
| 77 |
+
"learning_rate": 4.7324772605671486e-05,
|
| 78 |
+
"loss": 0.6543,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.17656500802568217,
|
| 83 |
+
"grad_norm": 1.4154126644134521,
|
| 84 |
+
"learning_rate": 4.7057249866238635e-05,
|
| 85 |
+
"loss": 0.6295,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1926163723916533,
|
| 90 |
+
"grad_norm": 1.4071663618087769,
|
| 91 |
+
"learning_rate": 4.678972712680578e-05,
|
| 92 |
+
"loss": 0.6088,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2086677367576244,
|
| 97 |
+
"grad_norm": 1.5164605379104614,
|
| 98 |
+
"learning_rate": 4.652220438737293e-05,
|
| 99 |
+
"loss": 0.6086,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.2247191011235955,
|
| 104 |
+
"grad_norm": 1.1694097518920898,
|
| 105 |
+
"learning_rate": 4.625468164794008e-05,
|
| 106 |
+
"loss": 0.6131,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.24077046548956663,
|
| 111 |
+
"grad_norm": 2.416905164718628,
|
| 112 |
+
"learning_rate": 4.598715890850723e-05,
|
| 113 |
+
"loss": 0.6392,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2568218298555377,
|
| 118 |
+
"grad_norm": 1.8865312337875366,
|
| 119 |
+
"learning_rate": 4.571963616907438e-05,
|
| 120 |
+
"loss": 0.5955,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.27287319422150885,
|
| 125 |
+
"grad_norm": 1.4470267295837402,
|
| 126 |
+
"learning_rate": 4.5452113429641525e-05,
|
| 127 |
+
"loss": 0.6245,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.28892455858747995,
|
| 132 |
+
"grad_norm": 2.124974489212036,
|
| 133 |
+
"learning_rate": 4.5184590690208673e-05,
|
| 134 |
+
"loss": 0.6251,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.30497592295345105,
|
| 139 |
+
"grad_norm": 2.1896016597747803,
|
| 140 |
+
"learning_rate": 4.491706795077582e-05,
|
| 141 |
+
"loss": 0.631,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.32102728731942215,
|
| 146 |
+
"grad_norm": 1.449318766593933,
|
| 147 |
+
"learning_rate": 4.4649545211342963e-05,
|
| 148 |
+
"loss": 0.5939,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.33707865168539325,
|
| 153 |
+
"grad_norm": 2.2219743728637695,
|
| 154 |
+
"learning_rate": 4.438202247191011e-05,
|
| 155 |
+
"loss": 0.5746,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.35313001605136435,
|
| 160 |
+
"grad_norm": 2.2409071922302246,
|
| 161 |
+
"learning_rate": 4.411449973247726e-05,
|
| 162 |
+
"loss": 0.6065,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.36918138041733545,
|
| 167 |
+
"grad_norm": 1.1223793029785156,
|
| 168 |
+
"learning_rate": 4.384697699304441e-05,
|
| 169 |
+
"loss": 0.6123,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3852327447833066,
|
| 174 |
+
"grad_norm": 1.2747622728347778,
|
| 175 |
+
"learning_rate": 4.357945425361156e-05,
|
| 176 |
+
"loss": 0.5952,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4012841091492777,
|
| 181 |
+
"grad_norm": 1.8558810949325562,
|
| 182 |
+
"learning_rate": 4.3311931514178705e-05,
|
| 183 |
+
"loss": 0.6027,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4173354735152488,
|
| 188 |
+
"grad_norm": 1.1629371643066406,
|
| 189 |
+
"learning_rate": 4.3044408774745854e-05,
|
| 190 |
+
"loss": 0.6228,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.4333868378812199,
|
| 195 |
+
"grad_norm": 1.4288586378097534,
|
| 196 |
+
"learning_rate": 4.2776886035313e-05,
|
| 197 |
+
"loss": 0.5734,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.449438202247191,
|
| 202 |
+
"grad_norm": 3.359997034072876,
|
| 203 |
+
"learning_rate": 4.250936329588015e-05,
|
| 204 |
+
"loss": 0.6018,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.4654895666131621,
|
| 209 |
+
"grad_norm": 1.8125280141830444,
|
| 210 |
+
"learning_rate": 4.22418405564473e-05,
|
| 211 |
+
"loss": 0.6288,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.48154093097913325,
|
| 216 |
+
"grad_norm": 1.3432456254959106,
|
| 217 |
+
"learning_rate": 4.197431781701445e-05,
|
| 218 |
+
"loss": 0.5756,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.49759229534510435,
|
| 223 |
+
"grad_norm": 3.372265100479126,
|
| 224 |
+
"learning_rate": 4.1706795077581596e-05,
|
| 225 |
+
"loss": 0.6197,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.5136436597110754,
|
| 230 |
+
"grad_norm": 1.7587655782699585,
|
| 231 |
+
"learning_rate": 4.1439272338148744e-05,
|
| 232 |
+
"loss": 0.5808,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5296950240770465,
|
| 237 |
+
"grad_norm": 2.1404621601104736,
|
| 238 |
+
"learning_rate": 4.117174959871589e-05,
|
| 239 |
+
"loss": 0.6075,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5457463884430177,
|
| 244 |
+
"grad_norm": 1.6256980895996094,
|
| 245 |
+
"learning_rate": 4.090422685928304e-05,
|
| 246 |
+
"loss": 0.5765,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5617977528089888,
|
| 251 |
+
"grad_norm": 2.2659549713134766,
|
| 252 |
+
"learning_rate": 4.063670411985019e-05,
|
| 253 |
+
"loss": 0.6305,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5778491171749599,
|
| 258 |
+
"grad_norm": 1.9907615184783936,
|
| 259 |
+
"learning_rate": 4.036918138041734e-05,
|
| 260 |
+
"loss": 0.5743,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.593900481540931,
|
| 265 |
+
"grad_norm": 2.26408314704895,
|
| 266 |
+
"learning_rate": 4.0101658640984486e-05,
|
| 267 |
+
"loss": 0.5842,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.6099518459069021,
|
| 272 |
+
"grad_norm": 1.9207652807235718,
|
| 273 |
+
"learning_rate": 3.9834135901551634e-05,
|
| 274 |
+
"loss": 0.5719,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.6260032102728732,
|
| 279 |
+
"grad_norm": 2.5978338718414307,
|
| 280 |
+
"learning_rate": 3.956661316211878e-05,
|
| 281 |
+
"loss": 0.5811,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.6420545746388443,
|
| 286 |
+
"grad_norm": 1.6617166996002197,
|
| 287 |
+
"learning_rate": 3.929909042268593e-05,
|
| 288 |
+
"loss": 0.5819,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6581059390048154,
|
| 293 |
+
"grad_norm": 2.7522661685943604,
|
| 294 |
+
"learning_rate": 3.903156768325308e-05,
|
| 295 |
+
"loss": 0.605,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.6741573033707865,
|
| 300 |
+
"grad_norm": 1.6527293920516968,
|
| 301 |
+
"learning_rate": 3.876404494382023e-05,
|
| 302 |
+
"loss": 0.6036,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.6902086677367576,
|
| 307 |
+
"grad_norm": 1.6553492546081543,
|
| 308 |
+
"learning_rate": 3.8496522204387376e-05,
|
| 309 |
+
"loss": 0.5723,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.7062600321027287,
|
| 314 |
+
"grad_norm": 1.9113073348999023,
|
| 315 |
+
"learning_rate": 3.8228999464954525e-05,
|
| 316 |
+
"loss": 0.5869,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.7223113964686998,
|
| 321 |
+
"grad_norm": 1.493342399597168,
|
| 322 |
+
"learning_rate": 3.796147672552167e-05,
|
| 323 |
+
"loss": 0.5912,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.7383627608346709,
|
| 328 |
+
"grad_norm": 1.3368749618530273,
|
| 329 |
+
"learning_rate": 3.769395398608882e-05,
|
| 330 |
+
"loss": 0.613,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.7544141252006421,
|
| 335 |
+
"grad_norm": 2.699831008911133,
|
| 336 |
+
"learning_rate": 3.742643124665597e-05,
|
| 337 |
+
"loss": 0.5658,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.7704654895666132,
|
| 342 |
+
"grad_norm": 1.0824522972106934,
|
| 343 |
+
"learning_rate": 3.715890850722312e-05,
|
| 344 |
+
"loss": 0.6116,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7865168539325843,
|
| 349 |
+
"grad_norm": 2.092763900756836,
|
| 350 |
+
"learning_rate": 3.689138576779027e-05,
|
| 351 |
+
"loss": 0.5711,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.8025682182985554,
|
| 356 |
+
"grad_norm": 2.2845699787139893,
|
| 357 |
+
"learning_rate": 3.6623863028357415e-05,
|
| 358 |
+
"loss": 0.5825,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.8186195826645265,
|
| 363 |
+
"grad_norm": 2.034006118774414,
|
| 364 |
+
"learning_rate": 3.6356340288924564e-05,
|
| 365 |
+
"loss": 0.5857,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.8346709470304976,
|
| 370 |
+
"grad_norm": 1.5809577703475952,
|
| 371 |
+
"learning_rate": 3.608881754949171e-05,
|
| 372 |
+
"loss": 0.5539,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.8507223113964687,
|
| 377 |
+
"grad_norm": 2.2914111614227295,
|
| 378 |
+
"learning_rate": 3.582129481005886e-05,
|
| 379 |
+
"loss": 0.5806,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.8667736757624398,
|
| 384 |
+
"grad_norm": 1.6067487001419067,
|
| 385 |
+
"learning_rate": 3.555377207062601e-05,
|
| 386 |
+
"loss": 0.5986,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.8828250401284109,
|
| 391 |
+
"grad_norm": 1.690928339958191,
|
| 392 |
+
"learning_rate": 3.528624933119316e-05,
|
| 393 |
+
"loss": 0.5713,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.898876404494382,
|
| 398 |
+
"grad_norm": 2.0696561336517334,
|
| 399 |
+
"learning_rate": 3.5018726591760305e-05,
|
| 400 |
+
"loss": 0.5601,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.9149277688603531,
|
| 405 |
+
"grad_norm": 1.693708896636963,
|
| 406 |
+
"learning_rate": 3.4751203852327454e-05,
|
| 407 |
+
"loss": 0.5812,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.9309791332263242,
|
| 412 |
+
"grad_norm": 1.5981098413467407,
|
| 413 |
+
"learning_rate": 3.44836811128946e-05,
|
| 414 |
+
"loss": 0.5699,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.9470304975922953,
|
| 419 |
+
"grad_norm": 1.4630780220031738,
|
| 420 |
+
"learning_rate": 3.421615837346175e-05,
|
| 421 |
+
"loss": 0.5917,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.9630818619582665,
|
| 426 |
+
"grad_norm": 1.8144547939300537,
|
| 427 |
+
"learning_rate": 3.394863563402889e-05,
|
| 428 |
+
"loss": 0.5969,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.9791332263242376,
|
| 433 |
+
"grad_norm": 2.242295265197754,
|
| 434 |
+
"learning_rate": 3.368111289459604e-05,
|
| 435 |
+
"loss": 0.5524,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.9951845906902087,
|
| 440 |
+
"grad_norm": 3.1214287281036377,
|
| 441 |
+
"learning_rate": 3.341359015516319e-05,
|
| 442 |
+
"loss": 0.5872,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.0112359550561798,
|
| 447 |
+
"grad_norm": 1.6238343715667725,
|
| 448 |
+
"learning_rate": 3.314606741573034e-05,
|
| 449 |
+
"loss": 0.5715,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.0272873194221508,
|
| 454 |
+
"grad_norm": 1.7471458911895752,
|
| 455 |
+
"learning_rate": 3.2878544676297486e-05,
|
| 456 |
+
"loss": 0.5377,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.043338683788122,
|
| 461 |
+
"grad_norm": 2.1529836654663086,
|
| 462 |
+
"learning_rate": 3.2611021936864634e-05,
|
| 463 |
+
"loss": 0.5542,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.0593900481540932,
|
| 468 |
+
"grad_norm": 2.278169870376587,
|
| 469 |
+
"learning_rate": 3.234349919743178e-05,
|
| 470 |
+
"loss": 0.5616,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.0754414125200642,
|
| 475 |
+
"grad_norm": 2.8219568729400635,
|
| 476 |
+
"learning_rate": 3.207597645799893e-05,
|
| 477 |
+
"loss": 0.5711,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.0914927768860354,
|
| 482 |
+
"grad_norm": 1.5639195442199707,
|
| 483 |
+
"learning_rate": 3.180845371856608e-05,
|
| 484 |
+
"loss": 0.5439,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.1075441412520064,
|
| 489 |
+
"grad_norm": 2.146303415298462,
|
| 490 |
+
"learning_rate": 3.154093097913323e-05,
|
| 491 |
+
"loss": 0.525,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.1235955056179776,
|
| 496 |
+
"grad_norm": 1.7315692901611328,
|
| 497 |
+
"learning_rate": 3.1273408239700376e-05,
|
| 498 |
+
"loss": 0.5521,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.1396468699839486,
|
| 503 |
+
"grad_norm": 2.457808494567871,
|
| 504 |
+
"learning_rate": 3.1005885500267525e-05,
|
| 505 |
+
"loss": 0.5209,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.1556982343499198,
|
| 510 |
+
"grad_norm": 2.4021546840667725,
|
| 511 |
+
"learning_rate": 3.073836276083467e-05,
|
| 512 |
+
"loss": 0.5056,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.1717495987158908,
|
| 517 |
+
"grad_norm": 3.025860548019409,
|
| 518 |
+
"learning_rate": 3.047084002140182e-05,
|
| 519 |
+
"loss": 0.4884,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.187800963081862,
|
| 524 |
+
"grad_norm": 3.1893551349639893,
|
| 525 |
+
"learning_rate": 3.0203317281968966e-05,
|
| 526 |
+
"loss": 0.5203,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.203852327447833,
|
| 531 |
+
"grad_norm": 4.527679443359375,
|
| 532 |
+
"learning_rate": 2.9935794542536115e-05,
|
| 533 |
+
"loss": 0.5114,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.2199036918138042,
|
| 538 |
+
"grad_norm": 3.021358013153076,
|
| 539 |
+
"learning_rate": 2.9668271803103263e-05,
|
| 540 |
+
"loss": 0.5408,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.2359550561797752,
|
| 545 |
+
"grad_norm": 3.0941548347473145,
|
| 546 |
+
"learning_rate": 2.940074906367041e-05,
|
| 547 |
+
"loss": 0.5122,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.2520064205457464,
|
| 552 |
+
"grad_norm": 2.5667285919189453,
|
| 553 |
+
"learning_rate": 2.913322632423756e-05,
|
| 554 |
+
"loss": 0.5192,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.2680577849117176,
|
| 559 |
+
"grad_norm": 3.7231733798980713,
|
| 560 |
+
"learning_rate": 2.886570358480471e-05,
|
| 561 |
+
"loss": 0.5609,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.2841091492776886,
|
| 566 |
+
"grad_norm": 1.9197113513946533,
|
| 567 |
+
"learning_rate": 2.8598180845371857e-05,
|
| 568 |
+
"loss": 0.4747,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.3001605136436596,
|
| 573 |
+
"grad_norm": 2.612793207168579,
|
| 574 |
+
"learning_rate": 2.8330658105939005e-05,
|
| 575 |
+
"loss": 0.5286,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.3162118780096308,
|
| 580 |
+
"grad_norm": 2.353598117828369,
|
| 581 |
+
"learning_rate": 2.8063135366506153e-05,
|
| 582 |
+
"loss": 0.535,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.332263242375602,
|
| 587 |
+
"grad_norm": 2.9524009227752686,
|
| 588 |
+
"learning_rate": 2.7795612627073302e-05,
|
| 589 |
+
"loss": 0.5402,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.348314606741573,
|
| 594 |
+
"grad_norm": 2.1400082111358643,
|
| 595 |
+
"learning_rate": 2.752808988764045e-05,
|
| 596 |
+
"loss": 0.5487,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.3643659711075442,
|
| 601 |
+
"grad_norm": 2.116978168487549,
|
| 602 |
+
"learning_rate": 2.72605671482076e-05,
|
| 603 |
+
"loss": 0.5586,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.3804173354735152,
|
| 608 |
+
"grad_norm": 2.3243775367736816,
|
| 609 |
+
"learning_rate": 2.6993044408774747e-05,
|
| 610 |
+
"loss": 0.5402,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.3964686998394864,
|
| 615 |
+
"grad_norm": 2.2850890159606934,
|
| 616 |
+
"learning_rate": 2.6725521669341895e-05,
|
| 617 |
+
"loss": 0.5151,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.4125200642054574,
|
| 622 |
+
"grad_norm": 2.305981397628784,
|
| 623 |
+
"learning_rate": 2.6457998929909044e-05,
|
| 624 |
+
"loss": 0.5494,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.4285714285714286,
|
| 629 |
+
"grad_norm": 1.8870021104812622,
|
| 630 |
+
"learning_rate": 2.6190476190476192e-05,
|
| 631 |
+
"loss": 0.5545,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.4446227929373996,
|
| 636 |
+
"grad_norm": 2.1851470470428467,
|
| 637 |
+
"learning_rate": 2.592295345104334e-05,
|
| 638 |
+
"loss": 0.5205,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.4606741573033708,
|
| 643 |
+
"grad_norm": 2.7136600017547607,
|
| 644 |
+
"learning_rate": 2.565543071161049e-05,
|
| 645 |
+
"loss": 0.5427,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.476725521669342,
|
| 650 |
+
"grad_norm": 3.9425771236419678,
|
| 651 |
+
"learning_rate": 2.5387907972177637e-05,
|
| 652 |
+
"loss": 0.5209,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.492776886035313,
|
| 657 |
+
"grad_norm": 2.84690260887146,
|
| 658 |
+
"learning_rate": 2.5120385232744786e-05,
|
| 659 |
+
"loss": 0.5562,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.508828250401284,
|
| 664 |
+
"grad_norm": 2.375824213027954,
|
| 665 |
+
"learning_rate": 2.485286249331193e-05,
|
| 666 |
+
"loss": 0.521,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.5248796147672552,
|
| 671 |
+
"grad_norm": 2.241267681121826,
|
| 672 |
+
"learning_rate": 2.458533975387908e-05,
|
| 673 |
+
"loss": 0.5576,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.5409309791332264,
|
| 678 |
+
"grad_norm": 2.209796190261841,
|
| 679 |
+
"learning_rate": 2.4317817014446228e-05,
|
| 680 |
+
"loss": 0.5034,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.5569823434991974,
|
| 685 |
+
"grad_norm": 2.9751803874969482,
|
| 686 |
+
"learning_rate": 2.4050294275013376e-05,
|
| 687 |
+
"loss": 0.5222,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.5730337078651684,
|
| 692 |
+
"grad_norm": 3.5506584644317627,
|
| 693 |
+
"learning_rate": 2.3782771535580524e-05,
|
| 694 |
+
"loss": 0.5067,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.5890850722311396,
|
| 699 |
+
"grad_norm": 2.4530675411224365,
|
| 700 |
+
"learning_rate": 2.3515248796147673e-05,
|
| 701 |
+
"loss": 0.5262,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.6051364365971108,
|
| 706 |
+
"grad_norm": 2.639045476913452,
|
| 707 |
+
"learning_rate": 2.324772605671482e-05,
|
| 708 |
+
"loss": 0.5396,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.621187800963082,
|
| 713 |
+
"grad_norm": 3.676542043685913,
|
| 714 |
+
"learning_rate": 2.298020331728197e-05,
|
| 715 |
+
"loss": 0.5462,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.637239165329053,
|
| 720 |
+
"grad_norm": 2.813171148300171,
|
| 721 |
+
"learning_rate": 2.2712680577849118e-05,
|
| 722 |
+
"loss": 0.5792,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.653290529695024,
|
| 727 |
+
"grad_norm": 3.5937094688415527,
|
| 728 |
+
"learning_rate": 2.2445157838416266e-05,
|
| 729 |
+
"loss": 0.527,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.6693418940609952,
|
| 734 |
+
"grad_norm": 2.3738603591918945,
|
| 735 |
+
"learning_rate": 2.2177635098983415e-05,
|
| 736 |
+
"loss": 0.5431,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.6853932584269664,
|
| 741 |
+
"grad_norm": 2.7167625427246094,
|
| 742 |
+
"learning_rate": 2.1910112359550563e-05,
|
| 743 |
+
"loss": 0.5187,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.7014446227929374,
|
| 748 |
+
"grad_norm": 2.8609092235565186,
|
| 749 |
+
"learning_rate": 2.164258962011771e-05,
|
| 750 |
+
"loss": 0.5061,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.7174959871589084,
|
| 755 |
+
"grad_norm": 3.4374756813049316,
|
| 756 |
+
"learning_rate": 2.137506688068486e-05,
|
| 757 |
+
"loss": 0.5301,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.7335473515248796,
|
| 762 |
+
"grad_norm": 2.3880395889282227,
|
| 763 |
+
"learning_rate": 2.1107544141252008e-05,
|
| 764 |
+
"loss": 0.5572,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.7495987158908508,
|
| 769 |
+
"grad_norm": 2.2175981998443604,
|
| 770 |
+
"learning_rate": 2.0840021401819157e-05,
|
| 771 |
+
"loss": 0.4862,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.7656500802568218,
|
| 776 |
+
"grad_norm": 1.8051007986068726,
|
| 777 |
+
"learning_rate": 2.05724986623863e-05,
|
| 778 |
+
"loss": 0.5108,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.7817014446227928,
|
| 783 |
+
"grad_norm": 3.5168681144714355,
|
| 784 |
+
"learning_rate": 2.030497592295345e-05,
|
| 785 |
+
"loss": 0.538,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.797752808988764,
|
| 790 |
+
"grad_norm": 2.5539281368255615,
|
| 791 |
+
"learning_rate": 2.00374531835206e-05,
|
| 792 |
+
"loss": 0.5489,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.8138041733547352,
|
| 797 |
+
"grad_norm": 1.7118955850601196,
|
| 798 |
+
"learning_rate": 1.9769930444087747e-05,
|
| 799 |
+
"loss": 0.5242,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.8298555377207064,
|
| 804 |
+
"grad_norm": 2.2016992568969727,
|
| 805 |
+
"learning_rate": 1.9502407704654895e-05,
|
| 806 |
+
"loss": 0.5172,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.8459069020866774,
|
| 811 |
+
"grad_norm": 2.072165012359619,
|
| 812 |
+
"learning_rate": 1.9234884965222044e-05,
|
| 813 |
+
"loss": 0.5216,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.8619582664526484,
|
| 818 |
+
"grad_norm": 2.446287155151367,
|
| 819 |
+
"learning_rate": 1.8967362225789192e-05,
|
| 820 |
+
"loss": 0.5391,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.8780096308186196,
|
| 825 |
+
"grad_norm": 2.6297097206115723,
|
| 826 |
+
"learning_rate": 1.869983948635634e-05,
|
| 827 |
+
"loss": 0.5046,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.8940609951845908,
|
| 832 |
+
"grad_norm": 2.429002285003662,
|
| 833 |
+
"learning_rate": 1.843231674692349e-05,
|
| 834 |
+
"loss": 0.5081,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.9101123595505618,
|
| 839 |
+
"grad_norm": 2.9561805725097656,
|
| 840 |
+
"learning_rate": 1.8164794007490637e-05,
|
| 841 |
+
"loss": 0.5335,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.9261637239165328,
|
| 846 |
+
"grad_norm": 2.9521896839141846,
|
| 847 |
+
"learning_rate": 1.7897271268057786e-05,
|
| 848 |
+
"loss": 0.5307,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.942215088282504,
|
| 853 |
+
"grad_norm": 2.409397602081299,
|
| 854 |
+
"learning_rate": 1.7629748528624934e-05,
|
| 855 |
+
"loss": 0.5014,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.9582664526484752,
|
| 860 |
+
"grad_norm": 2.046233892440796,
|
| 861 |
+
"learning_rate": 1.7362225789192082e-05,
|
| 862 |
+
"loss": 0.5637,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.9743178170144462,
|
| 867 |
+
"grad_norm": 2.7884459495544434,
|
| 868 |
+
"learning_rate": 1.709470304975923e-05,
|
| 869 |
+
"loss": 0.5512,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.9903691813804172,
|
| 874 |
+
"grad_norm": 2.0197086334228516,
|
| 875 |
+
"learning_rate": 1.682718031032638e-05,
|
| 876 |
+
"loss": 0.5112,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.0064205457463884,
|
| 881 |
+
"grad_norm": 2.801968574523926,
|
| 882 |
+
"learning_rate": 1.6559657570893527e-05,
|
| 883 |
+
"loss": 0.5084,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.0224719101123596,
|
| 888 |
+
"grad_norm": 2.099778652191162,
|
| 889 |
+
"learning_rate": 1.6292134831460676e-05,
|
| 890 |
+
"loss": 0.4558,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.038523274478331,
|
| 895 |
+
"grad_norm": 3.728285789489746,
|
| 896 |
+
"learning_rate": 1.6024612092027824e-05,
|
| 897 |
+
"loss": 0.453,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.0545746388443016,
|
| 902 |
+
"grad_norm": 3.892178773880005,
|
| 903 |
+
"learning_rate": 1.5757089352594973e-05,
|
| 904 |
+
"loss": 0.4347,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.070626003210273,
|
| 909 |
+
"grad_norm": 3.3752293586730957,
|
| 910 |
+
"learning_rate": 1.548956661316212e-05,
|
| 911 |
+
"loss": 0.4382,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.086677367576244,
|
| 916 |
+
"grad_norm": 4.784174919128418,
|
| 917 |
+
"learning_rate": 1.5222043873729266e-05,
|
| 918 |
+
"loss": 0.4416,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.102728731942215,
|
| 923 |
+
"grad_norm": 2.877718448638916,
|
| 924 |
+
"learning_rate": 1.4954521134296414e-05,
|
| 925 |
+
"loss": 0.4402,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.1187800963081864,
|
| 930 |
+
"grad_norm": 3.0553367137908936,
|
| 931 |
+
"learning_rate": 1.4686998394863563e-05,
|
| 932 |
+
"loss": 0.4508,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.134831460674157,
|
| 937 |
+
"grad_norm": 3.8250510692596436,
|
| 938 |
+
"learning_rate": 1.4419475655430711e-05,
|
| 939 |
+
"loss": 0.4422,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.1508828250401284,
|
| 944 |
+
"grad_norm": 3.143554210662842,
|
| 945 |
+
"learning_rate": 1.415195291599786e-05,
|
| 946 |
+
"loss": 0.4572,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.1669341894060996,
|
| 951 |
+
"grad_norm": 1.8287140130996704,
|
| 952 |
+
"learning_rate": 1.3884430176565008e-05,
|
| 953 |
+
"loss": 0.4287,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.182985553772071,
|
| 958 |
+
"grad_norm": 5.451256275177002,
|
| 959 |
+
"learning_rate": 1.3616907437132156e-05,
|
| 960 |
+
"loss": 0.4481,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.1990369181380416,
|
| 965 |
+
"grad_norm": 4.044301986694336,
|
| 966 |
+
"learning_rate": 1.3349384697699305e-05,
|
| 967 |
+
"loss": 0.4448,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.215088282504013,
|
| 972 |
+
"grad_norm": 4.877999782562256,
|
| 973 |
+
"learning_rate": 1.3081861958266453e-05,
|
| 974 |
+
"loss": 0.4457,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.231139646869984,
|
| 979 |
+
"grad_norm": 4.929600238800049,
|
| 980 |
+
"learning_rate": 1.2814339218833602e-05,
|
| 981 |
+
"loss": 0.432,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.247191011235955,
|
| 986 |
+
"grad_norm": 3.4969890117645264,
|
| 987 |
+
"learning_rate": 1.254681647940075e-05,
|
| 988 |
+
"loss": 0.4387,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.263242375601926,
|
| 993 |
+
"grad_norm": 4.1156086921691895,
|
| 994 |
+
"learning_rate": 1.2279293739967898e-05,
|
| 995 |
+
"loss": 0.4406,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.279293739967897,
|
| 1000 |
+
"grad_norm": 4.0271148681640625,
|
| 1001 |
+
"learning_rate": 1.2011771000535047e-05,
|
| 1002 |
+
"loss": 0.4283,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.2953451043338684,
|
| 1007 |
+
"grad_norm": 3.6957242488861084,
|
| 1008 |
+
"learning_rate": 1.1744248261102195e-05,
|
| 1009 |
+
"loss": 0.4424,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.3113964686998396,
|
| 1014 |
+
"grad_norm": 3.1115617752075195,
|
| 1015 |
+
"learning_rate": 1.1476725521669344e-05,
|
| 1016 |
+
"loss": 0.4475,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.3274478330658104,
|
| 1021 |
+
"grad_norm": 4.844674587249756,
|
| 1022 |
+
"learning_rate": 1.120920278223649e-05,
|
| 1023 |
+
"loss": 0.4459,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.3434991974317816,
|
| 1028 |
+
"grad_norm": 2.712280750274658,
|
| 1029 |
+
"learning_rate": 1.0941680042803639e-05,
|
| 1030 |
+
"loss": 0.4219,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.359550561797753,
|
| 1035 |
+
"grad_norm": 2.9259140491485596,
|
| 1036 |
+
"learning_rate": 1.0674157303370787e-05,
|
| 1037 |
+
"loss": 0.4824,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.375601926163724,
|
| 1042 |
+
"grad_norm": 4.346639633178711,
|
| 1043 |
+
"learning_rate": 1.0406634563937935e-05,
|
| 1044 |
+
"loss": 0.4364,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.391653290529695,
|
| 1049 |
+
"grad_norm": 2.7212588787078857,
|
| 1050 |
+
"learning_rate": 1.0139111824505084e-05,
|
| 1051 |
+
"loss": 0.3956,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.407704654895666,
|
| 1056 |
+
"grad_norm": 4.124871253967285,
|
| 1057 |
+
"learning_rate": 9.871589085072232e-06,
|
| 1058 |
+
"loss": 0.483,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.423756019261637,
|
| 1063 |
+
"grad_norm": 4.835409641265869,
|
| 1064 |
+
"learning_rate": 9.60406634563938e-06,
|
| 1065 |
+
"loss": 0.4623,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.4398073836276084,
|
| 1070 |
+
"grad_norm": 2.9034523963928223,
|
| 1071 |
+
"learning_rate": 9.336543606206529e-06,
|
| 1072 |
+
"loss": 0.4208,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.4558587479935796,
|
| 1077 |
+
"grad_norm": 2.569786787033081,
|
| 1078 |
+
"learning_rate": 9.069020866773677e-06,
|
| 1079 |
+
"loss": 0.4141,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.4719101123595504,
|
| 1084 |
+
"grad_norm": 4.244718551635742,
|
| 1085 |
+
"learning_rate": 8.801498127340826e-06,
|
| 1086 |
+
"loss": 0.4583,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.4879614767255216,
|
| 1091 |
+
"grad_norm": 4.004569053649902,
|
| 1092 |
+
"learning_rate": 8.533975387907972e-06,
|
| 1093 |
+
"loss": 0.4266,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.504012841091493,
|
| 1098 |
+
"grad_norm": 4.286050796508789,
|
| 1099 |
+
"learning_rate": 8.26645264847512e-06,
|
| 1100 |
+
"loss": 0.4117,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.520064205457464,
|
| 1105 |
+
"grad_norm": 4.992115497589111,
|
| 1106 |
+
"learning_rate": 7.99892990904227e-06,
|
| 1107 |
+
"loss": 0.4103,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.5361155698234352,
|
| 1112 |
+
"grad_norm": 4.154786586761475,
|
| 1113 |
+
"learning_rate": 7.731407169609418e-06,
|
| 1114 |
+
"loss": 0.4566,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.552166934189406,
|
| 1119 |
+
"grad_norm": 4.701552391052246,
|
| 1120 |
+
"learning_rate": 7.463884430176565e-06,
|
| 1121 |
+
"loss": 0.4534,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.568218298555377,
|
| 1126 |
+
"grad_norm": 3.9032320976257324,
|
| 1127 |
+
"learning_rate": 7.1963616907437135e-06,
|
| 1128 |
+
"loss": 0.4586,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.5842696629213484,
|
| 1133 |
+
"grad_norm": 4.401456832885742,
|
| 1134 |
+
"learning_rate": 6.928838951310862e-06,
|
| 1135 |
+
"loss": 0.4219,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.600321027287319,
|
| 1140 |
+
"grad_norm": 3.317080497741699,
|
| 1141 |
+
"learning_rate": 6.66131621187801e-06,
|
| 1142 |
+
"loss": 0.4522,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.6163723916532904,
|
| 1147 |
+
"grad_norm": 3.893983840942383,
|
| 1148 |
+
"learning_rate": 6.393793472445159e-06,
|
| 1149 |
+
"loss": 0.4208,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.6324237560192616,
|
| 1154 |
+
"grad_norm": 2.585857391357422,
|
| 1155 |
+
"learning_rate": 6.126270733012306e-06,
|
| 1156 |
+
"loss": 0.4535,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.648475120385233,
|
| 1161 |
+
"grad_norm": 2.956127405166626,
|
| 1162 |
+
"learning_rate": 5.858747993579455e-06,
|
| 1163 |
+
"loss": 0.4587,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.664526484751204,
|
| 1168 |
+
"grad_norm": 4.7360992431640625,
|
| 1169 |
+
"learning_rate": 5.591225254146603e-06,
|
| 1170 |
+
"loss": 0.4317,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.6805778491171752,
|
| 1175 |
+
"grad_norm": 3.546750068664551,
|
| 1176 |
+
"learning_rate": 5.323702514713751e-06,
|
| 1177 |
+
"loss": 0.4707,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.696629213483146,
|
| 1182 |
+
"grad_norm": 3.985381841659546,
|
| 1183 |
+
"learning_rate": 5.056179775280899e-06,
|
| 1184 |
+
"loss": 0.448,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.712680577849117,
|
| 1189 |
+
"grad_norm": 3.1598966121673584,
|
| 1190 |
+
"learning_rate": 4.788657035848047e-06,
|
| 1191 |
+
"loss": 0.4656,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.7287319422150884,
|
| 1196 |
+
"grad_norm": 2.8233871459960938,
|
| 1197 |
+
"learning_rate": 4.521134296415196e-06,
|
| 1198 |
+
"loss": 0.4827,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 2.744783306581059,
|
| 1203 |
+
"grad_norm": 3.8089215755462646,
|
| 1204 |
+
"learning_rate": 4.253611556982344e-06,
|
| 1205 |
+
"loss": 0.4454,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 2.7608346709470304,
|
| 1210 |
+
"grad_norm": 3.793998956680298,
|
| 1211 |
+
"learning_rate": 3.986088817549492e-06,
|
| 1212 |
+
"loss": 0.4626,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.7768860353130016,
|
| 1217 |
+
"grad_norm": 4.2339768409729,
|
| 1218 |
+
"learning_rate": 3.71856607811664e-06,
|
| 1219 |
+
"loss": 0.4368,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.792937399678973,
|
| 1224 |
+
"grad_norm": 3.96157169342041,
|
| 1225 |
+
"learning_rate": 3.4510433386837885e-06,
|
| 1226 |
+
"loss": 0.4189,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.808988764044944,
|
| 1231 |
+
"grad_norm": 3.535388708114624,
|
| 1232 |
+
"learning_rate": 3.1835205992509364e-06,
|
| 1233 |
+
"loss": 0.4585,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.825040128410915,
|
| 1238 |
+
"grad_norm": 3.344831943511963,
|
| 1239 |
+
"learning_rate": 2.9159978598180844e-06,
|
| 1240 |
+
"loss": 0.4366,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.841091492776886,
|
| 1245 |
+
"grad_norm": 2.6424927711486816,
|
| 1246 |
+
"learning_rate": 2.648475120385233e-06,
|
| 1247 |
+
"loss": 0.4465,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.857142857142857,
|
| 1252 |
+
"grad_norm": 3.244215250015259,
|
| 1253 |
+
"learning_rate": 2.3809523809523808e-06,
|
| 1254 |
+
"loss": 0.4404,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 2.8731942215088284,
|
| 1259 |
+
"grad_norm": 3.6791014671325684,
|
| 1260 |
+
"learning_rate": 2.113429641519529e-06,
|
| 1261 |
+
"loss": 0.4533,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.889245585874799,
|
| 1266 |
+
"grad_norm": 6.856778144836426,
|
| 1267 |
+
"learning_rate": 1.8459069020866775e-06,
|
| 1268 |
+
"loss": 0.4305,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.9052969502407704,
|
| 1273 |
+
"grad_norm": 5.275630474090576,
|
| 1274 |
+
"learning_rate": 1.5783841626538255e-06,
|
| 1275 |
+
"loss": 0.5065,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.9213483146067416,
|
| 1280 |
+
"grad_norm": 4.3441619873046875,
|
| 1281 |
+
"learning_rate": 1.310861423220974e-06,
|
| 1282 |
+
"loss": 0.3924,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 2.937399678972713,
|
| 1287 |
+
"grad_norm": 4.840381145477295,
|
| 1288 |
+
"learning_rate": 1.043338683788122e-06,
|
| 1289 |
+
"loss": 0.455,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 2.953451043338684,
|
| 1294 |
+
"grad_norm": 4.567574501037598,
|
| 1295 |
+
"learning_rate": 7.758159443552703e-07,
|
| 1296 |
+
"loss": 0.4415,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 2.969502407704655,
|
| 1301 |
+
"grad_norm": 3.6461331844329834,
|
| 1302 |
+
"learning_rate": 5.082932049224184e-07,
|
| 1303 |
+
"loss": 0.4363,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 2.985553772070626,
|
| 1308 |
+
"grad_norm": 5.579707145690918,
|
| 1309 |
+
"learning_rate": 2.407704654895666e-07,
|
| 1310 |
+
"loss": 0.4153,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
}
|
| 1313 |
+
],
|
| 1314 |
+
"logging_steps": 10,
|
| 1315 |
+
"max_steps": 1869,
|
| 1316 |
+
"num_input_tokens_seen": 0,
|
| 1317 |
+
"num_train_epochs": 3,
|
| 1318 |
+
"save_steps": 500,
|
| 1319 |
+
"stateful_callbacks": {
|
| 1320 |
+
"TrainerControl": {
|
| 1321 |
+
"args": {
|
| 1322 |
+
"should_epoch_stop": false,
|
| 1323 |
+
"should_evaluate": false,
|
| 1324 |
+
"should_log": false,
|
| 1325 |
+
"should_save": true,
|
| 1326 |
+
"should_training_stop": true
|
| 1327 |
+
},
|
| 1328 |
+
"attributes": {}
|
| 1329 |
+
}
|
| 1330 |
+
},
|
| 1331 |
+
"total_flos": 4324368475705500.0,
|
| 1332 |
+
"train_batch_size": 50,
|
| 1333 |
+
"trial_name": null,
|
| 1334 |
+
"trial_params": null
|
| 1335 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:612d2bd89977e65bf1d2b457eb4f3f54fcb926b5d64008e3b7fe90998ed90958
|
| 3 |
+
size 5048
|