رفع النموذج النهائي مع جميع الملفات للتجربة أو إعادة التدريب
Browse files- config.json +39 -0
- generation_config.json +9 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1824 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"pad_token_id": 50257,
|
| 21 |
+
"reorder_and_upcast_attn": false,
|
| 22 |
+
"resid_pdrop": 0.1,
|
| 23 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 24 |
+
"scale_attn_weights": true,
|
| 25 |
+
"summary_activation": null,
|
| 26 |
+
"summary_first_dropout": 0.1,
|
| 27 |
+
"summary_proj_to_labels": true,
|
| 28 |
+
"summary_type": "cls_index",
|
| 29 |
+
"summary_use_proj": true,
|
| 30 |
+
"task_specific_params": {
|
| 31 |
+
"text-generation": {
|
| 32 |
+
"do_sample": true,
|
| 33 |
+
"max_length": 50
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"transformers_version": "4.56.0",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50262
|
| 39 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
50256
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 50257,
|
| 8 |
+
"transformers_version": "4.56.0"
|
| 9 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b83ee74cfa07d3b414bfffd048b082931857d32a3d96cb64f808aca2906ad04
|
| 3 |
+
size 652194640
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6528b1a029139fd0f43aaf0fcbeedf5b12f24ec37e57ef5973ba35cee28e6c2
|
| 3 |
+
size 995669323
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af5350c323216b54d12bc98aa8d715adb26ea8f0465c9354173a7e6ba511c853
|
| 3 |
+
size 14455
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:450841536ed7378820c371e8ba97e31de7afbc122b03a7ea1b8daf67737a2475
|
| 3 |
+
size 1465
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1824 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 2000,
|
| 3 |
+
"best_metric": 0.8785109519958496,
|
| 4 |
+
"best_model_checkpoint": "./results/checkpoint-2000",
|
| 5 |
+
"epoch": 0.04941590401454804,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00019766361605819217,
|
| 14 |
+
"grad_norm": 12.468854904174805,
|
| 15 |
+
"learning_rate": 9e-07,
|
| 16 |
+
"loss": 3.6886,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00039532723211638433,
|
| 21 |
+
"grad_norm": 52.02771759033203,
|
| 22 |
+
"learning_rate": 1.9e-06,
|
| 23 |
+
"loss": 3.597,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0005929908481745765,
|
| 28 |
+
"grad_norm": 27.688196182250977,
|
| 29 |
+
"learning_rate": 2.9e-06,
|
| 30 |
+
"loss": 3.8032,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0007906544642327687,
|
| 35 |
+
"grad_norm": 12.250176429748535,
|
| 36 |
+
"learning_rate": 3.9e-06,
|
| 37 |
+
"loss": 3.3572,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.000988318080290961,
|
| 42 |
+
"grad_norm": 20.915935516357422,
|
| 43 |
+
"learning_rate": 4.9000000000000005e-06,
|
| 44 |
+
"loss": 3.3315,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.001185981696349153,
|
| 49 |
+
"grad_norm": 11.775025367736816,
|
| 50 |
+
"learning_rate": 5.9e-06,
|
| 51 |
+
"loss": 2.7832,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0013836453124073452,
|
| 56 |
+
"grad_norm": 12.706207275390625,
|
| 57 |
+
"learning_rate": 6.900000000000001e-06,
|
| 58 |
+
"loss": 2.3718,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.0015813089284655373,
|
| 63 |
+
"grad_norm": 21.651674270629883,
|
| 64 |
+
"learning_rate": 7.9e-06,
|
| 65 |
+
"loss": 2.0937,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0017789725445237295,
|
| 70 |
+
"grad_norm": 15.798470497131348,
|
| 71 |
+
"learning_rate": 8.9e-06,
|
| 72 |
+
"loss": 1.7039,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.001976636160581922,
|
| 77 |
+
"grad_norm": 9.725250244140625,
|
| 78 |
+
"learning_rate": 9.900000000000002e-06,
|
| 79 |
+
"loss": 1.5623,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.002174299776640114,
|
| 84 |
+
"grad_norm": 3.9574215412139893,
|
| 85 |
+
"learning_rate": 1.09e-05,
|
| 86 |
+
"loss": 1.6024,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.002371963392698306,
|
| 91 |
+
"grad_norm": 8.425467491149902,
|
| 92 |
+
"learning_rate": 1.19e-05,
|
| 93 |
+
"loss": 1.0913,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.0025696270087564982,
|
| 98 |
+
"grad_norm": 9.648313522338867,
|
| 99 |
+
"learning_rate": 1.29e-05,
|
| 100 |
+
"loss": 1.3395,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.0027672906248146904,
|
| 105 |
+
"grad_norm": 0.9009546041488647,
|
| 106 |
+
"learning_rate": 1.3900000000000002e-05,
|
| 107 |
+
"loss": 0.999,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0029649542408728825,
|
| 112 |
+
"grad_norm": 10.469375610351562,
|
| 113 |
+
"learning_rate": 1.49e-05,
|
| 114 |
+
"loss": 0.8274,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.0031626178569310747,
|
| 119 |
+
"grad_norm": 0.22269444167613983,
|
| 120 |
+
"learning_rate": 1.59e-05,
|
| 121 |
+
"loss": 0.8564,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.003360281472989267,
|
| 126 |
+
"grad_norm": 7.279628276824951,
|
| 127 |
+
"learning_rate": 1.69e-05,
|
| 128 |
+
"loss": 1.0921,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.003557945089047459,
|
| 133 |
+
"grad_norm": 0.6602649092674255,
|
| 134 |
+
"learning_rate": 1.79e-05,
|
| 135 |
+
"loss": 0.9095,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.003755608705105651,
|
| 140 |
+
"grad_norm": 0.1962263137102127,
|
| 141 |
+
"learning_rate": 1.8900000000000002e-05,
|
| 142 |
+
"loss": 0.6642,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.003953272321163844,
|
| 147 |
+
"grad_norm": 0.7642993927001953,
|
| 148 |
+
"learning_rate": 1.9900000000000003e-05,
|
| 149 |
+
"loss": 0.8014,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.004150935937222036,
|
| 154 |
+
"grad_norm": 0.7219191193580627,
|
| 155 |
+
"learning_rate": 2.09e-05,
|
| 156 |
+
"loss": 1.0484,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.004348599553280228,
|
| 161 |
+
"grad_norm": 7.698592185974121,
|
| 162 |
+
"learning_rate": 2.19e-05,
|
| 163 |
+
"loss": 1.2609,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.00454626316933842,
|
| 168 |
+
"grad_norm": 0.9080265760421753,
|
| 169 |
+
"learning_rate": 2.29e-05,
|
| 170 |
+
"loss": 0.9157,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.004743926785396612,
|
| 175 |
+
"grad_norm": 0.198422372341156,
|
| 176 |
+
"learning_rate": 2.39e-05,
|
| 177 |
+
"loss": 1.2826,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.004941590401454804,
|
| 182 |
+
"grad_norm": 0.267986923456192,
|
| 183 |
+
"learning_rate": 2.4900000000000002e-05,
|
| 184 |
+
"loss": 1.0538,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0051392540175129965,
|
| 189 |
+
"grad_norm": 9.535926818847656,
|
| 190 |
+
"learning_rate": 2.5900000000000003e-05,
|
| 191 |
+
"loss": 0.8342,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.005336917633571189,
|
| 196 |
+
"grad_norm": 0.20855465531349182,
|
| 197 |
+
"learning_rate": 2.6900000000000003e-05,
|
| 198 |
+
"loss": 0.4623,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.005534581249629381,
|
| 203 |
+
"grad_norm": 0.4157664477825165,
|
| 204 |
+
"learning_rate": 2.7900000000000004e-05,
|
| 205 |
+
"loss": 0.5038,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.005732244865687573,
|
| 210 |
+
"grad_norm": 8.263617515563965,
|
| 211 |
+
"learning_rate": 2.8899999999999998e-05,
|
| 212 |
+
"loss": 0.8435,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.005929908481745765,
|
| 217 |
+
"grad_norm": 9.216032981872559,
|
| 218 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 219 |
+
"loss": 0.7289,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.006127572097803957,
|
| 224 |
+
"grad_norm": 8.40715503692627,
|
| 225 |
+
"learning_rate": 3.09e-05,
|
| 226 |
+
"loss": 1.1021,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.006325235713862149,
|
| 231 |
+
"grad_norm": 0.4089910089969635,
|
| 232 |
+
"learning_rate": 3.19e-05,
|
| 233 |
+
"loss": 0.7686,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0065228993299203415,
|
| 238 |
+
"grad_norm": 8.227189064025879,
|
| 239 |
+
"learning_rate": 3.29e-05,
|
| 240 |
+
"loss": 0.9375,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.006720562945978534,
|
| 245 |
+
"grad_norm": 8.776185989379883,
|
| 246 |
+
"learning_rate": 3.3900000000000004e-05,
|
| 247 |
+
"loss": 0.9814,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.006918226562036726,
|
| 252 |
+
"grad_norm": 7.721053123474121,
|
| 253 |
+
"learning_rate": 3.49e-05,
|
| 254 |
+
"loss": 0.4118,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.007115890178094918,
|
| 259 |
+
"grad_norm": 5.676611423492432,
|
| 260 |
+
"learning_rate": 3.59e-05,
|
| 261 |
+
"loss": 1.1953,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.00731355379415311,
|
| 266 |
+
"grad_norm": 7.924537658691406,
|
| 267 |
+
"learning_rate": 3.69e-05,
|
| 268 |
+
"loss": 1.1895,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.007511217410211302,
|
| 273 |
+
"grad_norm": 8.677846908569336,
|
| 274 |
+
"learning_rate": 3.79e-05,
|
| 275 |
+
"loss": 0.8925,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.007708881026269494,
|
| 280 |
+
"grad_norm": 6.243840217590332,
|
| 281 |
+
"learning_rate": 3.8900000000000004e-05,
|
| 282 |
+
"loss": 0.6888,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.007906544642327687,
|
| 287 |
+
"grad_norm": 7.967802047729492,
|
| 288 |
+
"learning_rate": 3.99e-05,
|
| 289 |
+
"loss": 1.4355,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.008104208258385879,
|
| 294 |
+
"grad_norm": 8.38342571258545,
|
| 295 |
+
"learning_rate": 4.09e-05,
|
| 296 |
+
"loss": 1.0131,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.008301871874444072,
|
| 301 |
+
"grad_norm": 0.07280826568603516,
|
| 302 |
+
"learning_rate": 4.19e-05,
|
| 303 |
+
"loss": 1.3299,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.008499535490502263,
|
| 308 |
+
"grad_norm": 0.2540580630302429,
|
| 309 |
+
"learning_rate": 4.29e-05,
|
| 310 |
+
"loss": 0.6579,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.008697199106560456,
|
| 315 |
+
"grad_norm": 7.481705188751221,
|
| 316 |
+
"learning_rate": 4.39e-05,
|
| 317 |
+
"loss": 0.9635,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.008894862722618647,
|
| 322 |
+
"grad_norm": 0.20026662945747375,
|
| 323 |
+
"learning_rate": 4.49e-05,
|
| 324 |
+
"loss": 0.883,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.00909252633867684,
|
| 329 |
+
"grad_norm": 6.689612865447998,
|
| 330 |
+
"learning_rate": 4.5900000000000004e-05,
|
| 331 |
+
"loss": 0.9595,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.009290189954735031,
|
| 336 |
+
"grad_norm": 0.06714965403079987,
|
| 337 |
+
"learning_rate": 4.69e-05,
|
| 338 |
+
"loss": 0.7816,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.009487853570793224,
|
| 343 |
+
"grad_norm": 7.251598834991455,
|
| 344 |
+
"learning_rate": 4.79e-05,
|
| 345 |
+
"loss": 0.5981,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.009685517186851416,
|
| 350 |
+
"grad_norm": 6.720303058624268,
|
| 351 |
+
"learning_rate": 4.89e-05,
|
| 352 |
+
"loss": 0.9463,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.009883180802909609,
|
| 357 |
+
"grad_norm": 6.996522903442383,
|
| 358 |
+
"learning_rate": 4.99e-05,
|
| 359 |
+
"loss": 0.7919,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.009883180802909609,
|
| 364 |
+
"eval_loss": 0.9214622378349304,
|
| 365 |
+
"eval_runtime": 245.7679,
|
| 366 |
+
"eval_samples_per_second": 86.683,
|
| 367 |
+
"eval_steps_per_second": 10.835,
|
| 368 |
+
"step": 500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.0100808444189678,
|
| 372 |
+
"grad_norm": 0.3730940818786621,
|
| 373 |
+
"learning_rate": 4.9997025245747756e-05,
|
| 374 |
+
"loss": 0.8747,
|
| 375 |
+
"step": 510
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.010278508035025993,
|
| 379 |
+
"grad_norm": 6.384608268737793,
|
| 380 |
+
"learning_rate": 4.999371996324526e-05,
|
| 381 |
+
"loss": 1.1554,
|
| 382 |
+
"step": 520
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.010476171651084184,
|
| 386 |
+
"grad_norm": 6.133856773376465,
|
| 387 |
+
"learning_rate": 4.999041468074276e-05,
|
| 388 |
+
"loss": 1.1372,
|
| 389 |
+
"step": 530
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.010673835267142377,
|
| 393 |
+
"grad_norm": 0.2688162624835968,
|
| 394 |
+
"learning_rate": 4.998710939824027e-05,
|
| 395 |
+
"loss": 0.9546,
|
| 396 |
+
"step": 540
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.010871498883200569,
|
| 400 |
+
"grad_norm": 7.641685962677002,
|
| 401 |
+
"learning_rate": 4.9983804115737774e-05,
|
| 402 |
+
"loss": 0.9482,
|
| 403 |
+
"step": 550
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.011069162499258762,
|
| 407 |
+
"grad_norm": 6.708981037139893,
|
| 408 |
+
"learning_rate": 4.998049883323528e-05,
|
| 409 |
+
"loss": 0.6834,
|
| 410 |
+
"step": 560
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.011266826115316953,
|
| 414 |
+
"grad_norm": 7.363180637359619,
|
| 415 |
+
"learning_rate": 4.997719355073278e-05,
|
| 416 |
+
"loss": 0.7817,
|
| 417 |
+
"step": 570
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.011464489731375146,
|
| 421 |
+
"grad_norm": 6.942836284637451,
|
| 422 |
+
"learning_rate": 4.997388826823029e-05,
|
| 423 |
+
"loss": 1.0409,
|
| 424 |
+
"step": 580
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.011662153347433339,
|
| 428 |
+
"grad_norm": 5.417139530181885,
|
| 429 |
+
"learning_rate": 4.997058298572779e-05,
|
| 430 |
+
"loss": 1.1497,
|
| 431 |
+
"step": 590
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.01185981696349153,
|
| 435 |
+
"grad_norm": 0.11549051851034164,
|
| 436 |
+
"learning_rate": 4.9967277703225295e-05,
|
| 437 |
+
"loss": 0.671,
|
| 438 |
+
"step": 600
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.012057480579549723,
|
| 442 |
+
"grad_norm": 6.788494110107422,
|
| 443 |
+
"learning_rate": 4.99639724207228e-05,
|
| 444 |
+
"loss": 1.1224,
|
| 445 |
+
"step": 610
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.012255144195607914,
|
| 449 |
+
"grad_norm": 7.265660285949707,
|
| 450 |
+
"learning_rate": 4.996066713822031e-05,
|
| 451 |
+
"loss": 0.7574,
|
| 452 |
+
"step": 620
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.012452807811666107,
|
| 456 |
+
"grad_norm": 6.8422441482543945,
|
| 457 |
+
"learning_rate": 4.995736185571781e-05,
|
| 458 |
+
"loss": 0.668,
|
| 459 |
+
"step": 630
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.012650471427724299,
|
| 463 |
+
"grad_norm": 6.232938289642334,
|
| 464 |
+
"learning_rate": 4.995405657321531e-05,
|
| 465 |
+
"loss": 1.0247,
|
| 466 |
+
"step": 640
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.012848135043782492,
|
| 470 |
+
"grad_norm": 0.2170286625623703,
|
| 471 |
+
"learning_rate": 4.995075129071282e-05,
|
| 472 |
+
"loss": 0.8753,
|
| 473 |
+
"step": 650
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.013045798659840683,
|
| 477 |
+
"grad_norm": 5.531427383422852,
|
| 478 |
+
"learning_rate": 4.9947446008210326e-05,
|
| 479 |
+
"loss": 1.1531,
|
| 480 |
+
"step": 660
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.013243462275898876,
|
| 484 |
+
"grad_norm": 6.2044548988342285,
|
| 485 |
+
"learning_rate": 4.994414072570783e-05,
|
| 486 |
+
"loss": 0.8807,
|
| 487 |
+
"step": 670
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.013441125891957067,
|
| 491 |
+
"grad_norm": 6.0426836013793945,
|
| 492 |
+
"learning_rate": 4.994083544320533e-05,
|
| 493 |
+
"loss": 1.0564,
|
| 494 |
+
"step": 680
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.01363878950801526,
|
| 498 |
+
"grad_norm": 0.18100695312023163,
|
| 499 |
+
"learning_rate": 4.993753016070284e-05,
|
| 500 |
+
"loss": 0.845,
|
| 501 |
+
"step": 690
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.013836453124073451,
|
| 505 |
+
"grad_norm": 5.483642101287842,
|
| 506 |
+
"learning_rate": 4.9934224878200344e-05,
|
| 507 |
+
"loss": 1.1766,
|
| 508 |
+
"step": 700
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.014034116740131644,
|
| 512 |
+
"grad_norm": 4.922057628631592,
|
| 513 |
+
"learning_rate": 4.993091959569785e-05,
|
| 514 |
+
"loss": 1.0413,
|
| 515 |
+
"step": 710
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.014231780356189836,
|
| 519 |
+
"grad_norm": 5.475461959838867,
|
| 520 |
+
"learning_rate": 4.9927614313195356e-05,
|
| 521 |
+
"loss": 0.9466,
|
| 522 |
+
"step": 720
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.014429443972248029,
|
| 526 |
+
"grad_norm": 6.56538200378418,
|
| 527 |
+
"learning_rate": 4.992430903069286e-05,
|
| 528 |
+
"loss": 0.9341,
|
| 529 |
+
"step": 730
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.01462710758830622,
|
| 533 |
+
"grad_norm": 4.976065635681152,
|
| 534 |
+
"learning_rate": 4.992100374819036e-05,
|
| 535 |
+
"loss": 0.8329,
|
| 536 |
+
"step": 740
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.014824771204364413,
|
| 540 |
+
"grad_norm": 0.05167540907859802,
|
| 541 |
+
"learning_rate": 4.9917698465687865e-05,
|
| 542 |
+
"loss": 0.698,
|
| 543 |
+
"step": 750
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.015022434820422604,
|
| 547 |
+
"grad_norm": 0.05338999629020691,
|
| 548 |
+
"learning_rate": 4.991439318318537e-05,
|
| 549 |
+
"loss": 0.4403,
|
| 550 |
+
"step": 760
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.015220098436480797,
|
| 554 |
+
"grad_norm": 6.14265251159668,
|
| 555 |
+
"learning_rate": 4.991108790068287e-05,
|
| 556 |
+
"loss": 0.9864,
|
| 557 |
+
"step": 770
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.015417762052538989,
|
| 561 |
+
"grad_norm": 0.1925213485956192,
|
| 562 |
+
"learning_rate": 4.9907782618180374e-05,
|
| 563 |
+
"loss": 1.1037,
|
| 564 |
+
"step": 780
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.015615425668597182,
|
| 568 |
+
"grad_norm": 5.5837554931640625,
|
| 569 |
+
"learning_rate": 4.990447733567788e-05,
|
| 570 |
+
"loss": 0.8571,
|
| 571 |
+
"step": 790
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.015813089284655375,
|
| 575 |
+
"grad_norm": 3.6623053550720215,
|
| 576 |
+
"learning_rate": 4.9901172053175386e-05,
|
| 577 |
+
"loss": 0.9082,
|
| 578 |
+
"step": 800
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.016010752900713566,
|
| 582 |
+
"grad_norm": 0.05198395252227783,
|
| 583 |
+
"learning_rate": 4.989786677067289e-05,
|
| 584 |
+
"loss": 0.8335,
|
| 585 |
+
"step": 810
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.016208416516771757,
|
| 589 |
+
"grad_norm": 5.30173921585083,
|
| 590 |
+
"learning_rate": 4.989456148817039e-05,
|
| 591 |
+
"loss": 0.6003,
|
| 592 |
+
"step": 820
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.01640608013282995,
|
| 596 |
+
"grad_norm": 4.080887317657471,
|
| 597 |
+
"learning_rate": 4.98912562056679e-05,
|
| 598 |
+
"loss": 0.7937,
|
| 599 |
+
"step": 830
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.016603743748888143,
|
| 603 |
+
"grad_norm": 5.606449604034424,
|
| 604 |
+
"learning_rate": 4.9887950923165404e-05,
|
| 605 |
+
"loss": 1.2462,
|
| 606 |
+
"step": 840
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.016801407364946334,
|
| 610 |
+
"grad_norm": 6.573556900024414,
|
| 611 |
+
"learning_rate": 4.988464564066291e-05,
|
| 612 |
+
"loss": 0.819,
|
| 613 |
+
"step": 850
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.016999070981004526,
|
| 617 |
+
"grad_norm": 0.06031334027647972,
|
| 618 |
+
"learning_rate": 4.988134035816042e-05,
|
| 619 |
+
"loss": 0.6425,
|
| 620 |
+
"step": 860
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.017196734597062717,
|
| 624 |
+
"grad_norm": 4.252384662628174,
|
| 625 |
+
"learning_rate": 4.987803507565792e-05,
|
| 626 |
+
"loss": 0.8228,
|
| 627 |
+
"step": 870
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.01739439821312091,
|
| 631 |
+
"grad_norm": 0.12518936395645142,
|
| 632 |
+
"learning_rate": 4.987472979315542e-05,
|
| 633 |
+
"loss": 0.7576,
|
| 634 |
+
"step": 880
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.017592061829179103,
|
| 638 |
+
"grad_norm": 5.032276630401611,
|
| 639 |
+
"learning_rate": 4.9871424510652925e-05,
|
| 640 |
+
"loss": 0.8783,
|
| 641 |
+
"step": 890
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.017789725445237294,
|
| 645 |
+
"grad_norm": 0.059717316180467606,
|
| 646 |
+
"learning_rate": 4.9868119228150435e-05,
|
| 647 |
+
"loss": 0.8153,
|
| 648 |
+
"step": 900
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.01798738906129549,
|
| 652 |
+
"grad_norm": 3.093669891357422,
|
| 653 |
+
"learning_rate": 4.986481394564794e-05,
|
| 654 |
+
"loss": 0.6105,
|
| 655 |
+
"step": 910
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.01818505267735368,
|
| 659 |
+
"grad_norm": 3.593827962875366,
|
| 660 |
+
"learning_rate": 4.986150866314544e-05,
|
| 661 |
+
"loss": 0.6549,
|
| 662 |
+
"step": 920
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.01838271629341187,
|
| 666 |
+
"grad_norm": 0.11692095547914505,
|
| 667 |
+
"learning_rate": 4.985820338064295e-05,
|
| 668 |
+
"loss": 1.0544,
|
| 669 |
+
"step": 930
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.018580379909470063,
|
| 673 |
+
"grad_norm": 0.07805492728948593,
|
| 674 |
+
"learning_rate": 4.985489809814045e-05,
|
| 675 |
+
"loss": 0.8575,
|
| 676 |
+
"step": 940
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.018778043525528258,
|
| 680 |
+
"grad_norm": 3.621807098388672,
|
| 681 |
+
"learning_rate": 4.9851592815637956e-05,
|
| 682 |
+
"loss": 1.1574,
|
| 683 |
+
"step": 950
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.01897570714158645,
|
| 687 |
+
"grad_norm": 0.1467035412788391,
|
| 688 |
+
"learning_rate": 4.984828753313546e-05,
|
| 689 |
+
"loss": 0.9261,
|
| 690 |
+
"step": 960
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.01917337075764464,
|
| 694 |
+
"grad_norm": 5.1112895011901855,
|
| 695 |
+
"learning_rate": 4.984498225063297e-05,
|
| 696 |
+
"loss": 1.1392,
|
| 697 |
+
"step": 970
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.01937103437370283,
|
| 701 |
+
"grad_norm": 4.554998397827148,
|
| 702 |
+
"learning_rate": 4.984167696813047e-05,
|
| 703 |
+
"loss": 1.0693,
|
| 704 |
+
"step": 980
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.019568697989761026,
|
| 708 |
+
"grad_norm": 0.08779212087392807,
|
| 709 |
+
"learning_rate": 4.9838371685627974e-05,
|
| 710 |
+
"loss": 0.6191,
|
| 711 |
+
"step": 990
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.019766361605819217,
|
| 715 |
+
"grad_norm": 0.050688233226537704,
|
| 716 |
+
"learning_rate": 4.983506640312548e-05,
|
| 717 |
+
"loss": 0.9877,
|
| 718 |
+
"step": 1000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.019766361605819217,
|
| 722 |
+
"eval_loss": 0.9038341045379639,
|
| 723 |
+
"eval_runtime": 236.3116,
|
| 724 |
+
"eval_samples_per_second": 90.152,
|
| 725 |
+
"eval_steps_per_second": 11.269,
|
| 726 |
+
"step": 1000
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 0.01996402522187741,
|
| 730 |
+
"grad_norm": 0.0693553164601326,
|
| 731 |
+
"learning_rate": 4.983176112062298e-05,
|
| 732 |
+
"loss": 1.1317,
|
| 733 |
+
"step": 1010
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 0.0201616888379356,
|
| 737 |
+
"grad_norm": 4.626270771026611,
|
| 738 |
+
"learning_rate": 4.982845583812048e-05,
|
| 739 |
+
"loss": 0.9338,
|
| 740 |
+
"step": 1020
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.020359352453993795,
|
| 744 |
+
"grad_norm": 4.608981132507324,
|
| 745 |
+
"learning_rate": 4.9825150555617986e-05,
|
| 746 |
+
"loss": 1.0849,
|
| 747 |
+
"step": 1030
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.020557016070051986,
|
| 751 |
+
"grad_norm": 4.660639762878418,
|
| 752 |
+
"learning_rate": 4.9821845273115495e-05,
|
| 753 |
+
"loss": 1.4633,
|
| 754 |
+
"step": 1040
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.020754679686110177,
|
| 758 |
+
"grad_norm": 4.729643821716309,
|
| 759 |
+
"learning_rate": 4.9818539990613e-05,
|
| 760 |
+
"loss": 1.0748,
|
| 761 |
+
"step": 1050
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.02095234330216837,
|
| 765 |
+
"grad_norm": 0.027232196182012558,
|
| 766 |
+
"learning_rate": 4.98152347081105e-05,
|
| 767 |
+
"loss": 0.7562,
|
| 768 |
+
"step": 1060
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.021150006918226563,
|
| 772 |
+
"grad_norm": 0.018449297174811363,
|
| 773 |
+
"learning_rate": 4.981192942560801e-05,
|
| 774 |
+
"loss": 0.4423,
|
| 775 |
+
"step": 1070
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.021347670534284754,
|
| 779 |
+
"grad_norm": 3.372770309448242,
|
| 780 |
+
"learning_rate": 4.9808624143105514e-05,
|
| 781 |
+
"loss": 0.7371,
|
| 782 |
+
"step": 1080
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.021545334150342946,
|
| 786 |
+
"grad_norm": 0.24770286679267883,
|
| 787 |
+
"learning_rate": 4.9805318860603016e-05,
|
| 788 |
+
"loss": 1.4439,
|
| 789 |
+
"step": 1090
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.021742997766401137,
|
| 793 |
+
"grad_norm": 0.11222311854362488,
|
| 794 |
+
"learning_rate": 4.980201357810052e-05,
|
| 795 |
+
"loss": 0.6073,
|
| 796 |
+
"step": 1100
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.021940661382459332,
|
| 800 |
+
"grad_norm": 4.456133842468262,
|
| 801 |
+
"learning_rate": 4.979870829559803e-05,
|
| 802 |
+
"loss": 0.6697,
|
| 803 |
+
"step": 1110
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.022138324998517523,
|
| 807 |
+
"grad_norm": 0.05747058987617493,
|
| 808 |
+
"learning_rate": 4.979540301309553e-05,
|
| 809 |
+
"loss": 0.949,
|
| 810 |
+
"step": 1120
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.022335988614575714,
|
| 814 |
+
"grad_norm": 3.084164619445801,
|
| 815 |
+
"learning_rate": 4.9792097730593035e-05,
|
| 816 |
+
"loss": 0.8189,
|
| 817 |
+
"step": 1130
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.022533652230633906,
|
| 821 |
+
"grad_norm": 3.456510305404663,
|
| 822 |
+
"learning_rate": 4.9788792448090544e-05,
|
| 823 |
+
"loss": 1.3608,
|
| 824 |
+
"step": 1140
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.0227313158466921,
|
| 828 |
+
"grad_norm": 3.4595232009887695,
|
| 829 |
+
"learning_rate": 4.978548716558805e-05,
|
| 830 |
+
"loss": 0.8679,
|
| 831 |
+
"step": 1150
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.02292897946275029,
|
| 835 |
+
"grad_norm": 4.478796005249023,
|
| 836 |
+
"learning_rate": 4.978218188308555e-05,
|
| 837 |
+
"loss": 0.8229,
|
| 838 |
+
"step": 1160
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 0.023126643078808483,
|
| 842 |
+
"grad_norm": 0.07277049869298935,
|
| 843 |
+
"learning_rate": 4.977887660058305e-05,
|
| 844 |
+
"loss": 0.5847,
|
| 845 |
+
"step": 1170
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.023324306694866678,
|
| 849 |
+
"grad_norm": 0.13703665137290955,
|
| 850 |
+
"learning_rate": 4.977557131808056e-05,
|
| 851 |
+
"loss": 1.0406,
|
| 852 |
+
"step": 1180
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.02352197031092487,
|
| 856 |
+
"grad_norm": 3.7098865509033203,
|
| 857 |
+
"learning_rate": 4.9772266035578065e-05,
|
| 858 |
+
"loss": 0.4147,
|
| 859 |
+
"step": 1190
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.02371963392698306,
|
| 863 |
+
"grad_norm": 0.03374309837818146,
|
| 864 |
+
"learning_rate": 4.976896075307557e-05,
|
| 865 |
+
"loss": 0.8773,
|
| 866 |
+
"step": 1200
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.02391729754304125,
|
| 870 |
+
"grad_norm": 4.257831573486328,
|
| 871 |
+
"learning_rate": 4.976565547057308e-05,
|
| 872 |
+
"loss": 0.6989,
|
| 873 |
+
"step": 1210
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.024114961159099446,
|
| 877 |
+
"grad_norm": 4.520907402038574,
|
| 878 |
+
"learning_rate": 4.976235018807058e-05,
|
| 879 |
+
"loss": 0.6721,
|
| 880 |
+
"step": 1220
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.024312624775157637,
|
| 884 |
+
"grad_norm": 7.800019264221191,
|
| 885 |
+
"learning_rate": 4.9759044905568084e-05,
|
| 886 |
+
"loss": 0.9356,
|
| 887 |
+
"step": 1230
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.02451028839121583,
|
| 891 |
+
"grad_norm": 3.6012537479400635,
|
| 892 |
+
"learning_rate": 4.9755739623065586e-05,
|
| 893 |
+
"loss": 0.9297,
|
| 894 |
+
"step": 1240
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.02470795200727402,
|
| 898 |
+
"grad_norm": 4.273988246917725,
|
| 899 |
+
"learning_rate": 4.975243434056309e-05,
|
| 900 |
+
"loss": 1.1335,
|
| 901 |
+
"step": 1250
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.024905615623332215,
|
| 905 |
+
"grad_norm": 0.15219643712043762,
|
| 906 |
+
"learning_rate": 4.974912905806059e-05,
|
| 907 |
+
"loss": 0.608,
|
| 908 |
+
"step": 1260
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.025103279239390406,
|
| 912 |
+
"grad_norm": 4.302224636077881,
|
| 913 |
+
"learning_rate": 4.9745823775558095e-05,
|
| 914 |
+
"loss": 0.974,
|
| 915 |
+
"step": 1270
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.025300942855448597,
|
| 919 |
+
"grad_norm": 0.09040108323097229,
|
| 920 |
+
"learning_rate": 4.9742518493055605e-05,
|
| 921 |
+
"loss": 0.871,
|
| 922 |
+
"step": 1280
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.02549860647150679,
|
| 926 |
+
"grad_norm": 0.08080489933490753,
|
| 927 |
+
"learning_rate": 4.973921321055311e-05,
|
| 928 |
+
"loss": 1.0142,
|
| 929 |
+
"step": 1290
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.025696270087564983,
|
| 933 |
+
"grad_norm": 4.567043781280518,
|
| 934 |
+
"learning_rate": 4.973590792805061e-05,
|
| 935 |
+
"loss": 0.9593,
|
| 936 |
+
"step": 1300
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.025893933703623175,
|
| 940 |
+
"grad_norm": 3.6397976875305176,
|
| 941 |
+
"learning_rate": 4.973260264554811e-05,
|
| 942 |
+
"loss": 1.0869,
|
| 943 |
+
"step": 1310
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 0.026091597319681366,
|
| 947 |
+
"grad_norm": 3.9633944034576416,
|
| 948 |
+
"learning_rate": 4.972929736304562e-05,
|
| 949 |
+
"loss": 0.8078,
|
| 950 |
+
"step": 1320
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.026289260935739557,
|
| 954 |
+
"grad_norm": 2.7522361278533936,
|
| 955 |
+
"learning_rate": 4.9725992080543126e-05,
|
| 956 |
+
"loss": 1.0228,
|
| 957 |
+
"step": 1330
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 0.026486924551797752,
|
| 961 |
+
"grad_norm": 3.5321545600891113,
|
| 962 |
+
"learning_rate": 4.972268679804063e-05,
|
| 963 |
+
"loss": 1.1676,
|
| 964 |
+
"step": 1340
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 0.026684588167855943,
|
| 968 |
+
"grad_norm": 3.301154851913452,
|
| 969 |
+
"learning_rate": 4.971938151553814e-05,
|
| 970 |
+
"loss": 0.9431,
|
| 971 |
+
"step": 1350
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 0.026882251783914134,
|
| 975 |
+
"grad_norm": 0.0989571139216423,
|
| 976 |
+
"learning_rate": 4.971607623303564e-05,
|
| 977 |
+
"loss": 1.0831,
|
| 978 |
+
"step": 1360
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.027079915399972326,
|
| 982 |
+
"grad_norm": 3.8333899974823,
|
| 983 |
+
"learning_rate": 4.9712770950533144e-05,
|
| 984 |
+
"loss": 1.3761,
|
| 985 |
+
"step": 1370
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.02727757901603052,
|
| 989 |
+
"grad_norm": 3.214996099472046,
|
| 990 |
+
"learning_rate": 4.970946566803065e-05,
|
| 991 |
+
"loss": 0.7668,
|
| 992 |
+
"step": 1380
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.02747524263208871,
|
| 996 |
+
"grad_norm": 3.960110902786255,
|
| 997 |
+
"learning_rate": 4.9706160385528156e-05,
|
| 998 |
+
"loss": 1.002,
|
| 999 |
+
"step": 1390
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.027672906248146903,
|
| 1003 |
+
"grad_norm": 4.702463150024414,
|
| 1004 |
+
"learning_rate": 4.970285510302566e-05,
|
| 1005 |
+
"loss": 1.013,
|
| 1006 |
+
"step": 1400
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.027870569864205094,
|
| 1010 |
+
"grad_norm": 4.348130226135254,
|
| 1011 |
+
"learning_rate": 4.969954982052316e-05,
|
| 1012 |
+
"loss": 0.828,
|
| 1013 |
+
"step": 1410
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 0.02806823348026329,
|
| 1017 |
+
"grad_norm": 0.1650361716747284,
|
| 1018 |
+
"learning_rate": 4.9696244538020665e-05,
|
| 1019 |
+
"loss": 0.4713,
|
| 1020 |
+
"step": 1420
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 0.02826589709632148,
|
| 1024 |
+
"grad_norm": 0.0727657675743103,
|
| 1025 |
+
"learning_rate": 4.9692939255518175e-05,
|
| 1026 |
+
"loss": 0.9447,
|
| 1027 |
+
"step": 1430
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 0.02846356071237967,
|
| 1031 |
+
"grad_norm": 3.5271363258361816,
|
| 1032 |
+
"learning_rate": 4.968963397301568e-05,
|
| 1033 |
+
"loss": 1.0093,
|
| 1034 |
+
"step": 1440
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.028661224328437863,
|
| 1038 |
+
"grad_norm": 0.10453560948371887,
|
| 1039 |
+
"learning_rate": 4.968632869051318e-05,
|
| 1040 |
+
"loss": 0.7617,
|
| 1041 |
+
"step": 1450
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.028858887944496057,
|
| 1045 |
+
"grad_norm": 2.5674312114715576,
|
| 1046 |
+
"learning_rate": 4.968302340801069e-05,
|
| 1047 |
+
"loss": 1.0254,
|
| 1048 |
+
"step": 1460
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 0.02905655156055425,
|
| 1052 |
+
"grad_norm": 4.187676429748535,
|
| 1053 |
+
"learning_rate": 4.967971812550819e-05,
|
| 1054 |
+
"loss": 1.119,
|
| 1055 |
+
"step": 1470
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 0.02925421517661244,
|
| 1059 |
+
"grad_norm": 0.09666266292333603,
|
| 1060 |
+
"learning_rate": 4.9676412843005696e-05,
|
| 1061 |
+
"loss": 0.6603,
|
| 1062 |
+
"step": 1480
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 0.029451878792670635,
|
| 1066 |
+
"grad_norm": 3.545473098754883,
|
| 1067 |
+
"learning_rate": 4.96731075605032e-05,
|
| 1068 |
+
"loss": 1.0676,
|
| 1069 |
+
"step": 1490
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 0.029649542408728826,
|
| 1073 |
+
"grad_norm": 3.4705913066864014,
|
| 1074 |
+
"learning_rate": 4.96698022780007e-05,
|
| 1075 |
+
"loss": 0.6608,
|
| 1076 |
+
"step": 1500
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 0.029649542408728826,
|
| 1080 |
+
"eval_loss": 0.8907366394996643,
|
| 1081 |
+
"eval_runtime": 236.1612,
|
| 1082 |
+
"eval_samples_per_second": 90.21,
|
| 1083 |
+
"eval_steps_per_second": 11.276,
|
| 1084 |
+
"step": 1500
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 0.029847206024787017,
|
| 1088 |
+
"grad_norm": 0.06528951227664948,
|
| 1089 |
+
"learning_rate": 4.9666496995498204e-05,
|
| 1090 |
+
"loss": 0.8131,
|
| 1091 |
+
"step": 1510
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.03004486964084521,
|
| 1095 |
+
"grad_norm": 0.11044718325138092,
|
| 1096 |
+
"learning_rate": 4.966319171299571e-05,
|
| 1097 |
+
"loss": 0.9144,
|
| 1098 |
+
"step": 1520
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 0.030242533256903403,
|
| 1102 |
+
"grad_norm": 3.092930316925049,
|
| 1103 |
+
"learning_rate": 4.965988643049322e-05,
|
| 1104 |
+
"loss": 1.0863,
|
| 1105 |
+
"step": 1530
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.030440196872961595,
|
| 1109 |
+
"grad_norm": 3.9734549522399902,
|
| 1110 |
+
"learning_rate": 4.965658114799072e-05,
|
| 1111 |
+
"loss": 0.5064,
|
| 1112 |
+
"step": 1540
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.030637860489019786,
|
| 1116 |
+
"grad_norm": 0.035740889608860016,
|
| 1117 |
+
"learning_rate": 4.965327586548822e-05,
|
| 1118 |
+
"loss": 1.222,
|
| 1119 |
+
"step": 1550
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.030835524105077977,
|
| 1123 |
+
"grad_norm": 0.06910470873117447,
|
| 1124 |
+
"learning_rate": 4.9649970582985725e-05,
|
| 1125 |
+
"loss": 0.8428,
|
| 1126 |
+
"step": 1560
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 0.031033187721136172,
|
| 1130 |
+
"grad_norm": 3.261096715927124,
|
| 1131 |
+
"learning_rate": 4.9646665300483235e-05,
|
| 1132 |
+
"loss": 0.9363,
|
| 1133 |
+
"step": 1570
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 0.031230851337194363,
|
| 1137 |
+
"grad_norm": 0.09766729176044464,
|
| 1138 |
+
"learning_rate": 4.964336001798074e-05,
|
| 1139 |
+
"loss": 0.803,
|
| 1140 |
+
"step": 1580
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.03142851495325256,
|
| 1144 |
+
"grad_norm": 3.2366461753845215,
|
| 1145 |
+
"learning_rate": 4.964005473547824e-05,
|
| 1146 |
+
"loss": 1.2179,
|
| 1147 |
+
"step": 1590
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 0.03162617856931075,
|
| 1151 |
+
"grad_norm": 0.09953774511814117,
|
| 1152 |
+
"learning_rate": 4.963674945297575e-05,
|
| 1153 |
+
"loss": 1.0656,
|
| 1154 |
+
"step": 1600
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.03182384218536894,
|
| 1158 |
+
"grad_norm": 3.7796168327331543,
|
| 1159 |
+
"learning_rate": 4.963344417047325e-05,
|
| 1160 |
+
"loss": 1.0625,
|
| 1161 |
+
"step": 1610
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.03202150580142713,
|
| 1165 |
+
"grad_norm": 0.16226442158222198,
|
| 1166 |
+
"learning_rate": 4.9630138887970756e-05,
|
| 1167 |
+
"loss": 0.8619,
|
| 1168 |
+
"step": 1620
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.03221916941748532,
|
| 1172 |
+
"grad_norm": 3.7525415420532227,
|
| 1173 |
+
"learning_rate": 4.962683360546826e-05,
|
| 1174 |
+
"loss": 0.9852,
|
| 1175 |
+
"step": 1630
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 0.032416833033543514,
|
| 1179 |
+
"grad_norm": 0.07482007145881653,
|
| 1180 |
+
"learning_rate": 4.962352832296577e-05,
|
| 1181 |
+
"loss": 0.8297,
|
| 1182 |
+
"step": 1640
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.032614496649601706,
|
| 1186 |
+
"grad_norm": 0.1131066307425499,
|
| 1187 |
+
"learning_rate": 4.962022304046327e-05,
|
| 1188 |
+
"loss": 1.1424,
|
| 1189 |
+
"step": 1650
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 0.0328121602656599,
|
| 1193 |
+
"grad_norm": 0.05459194257855415,
|
| 1194 |
+
"learning_rate": 4.9616917757960774e-05,
|
| 1195 |
+
"loss": 0.7202,
|
| 1196 |
+
"step": 1660
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 0.033009823881718095,
|
| 1200 |
+
"grad_norm": 3.9901282787323,
|
| 1201 |
+
"learning_rate": 4.9613612475458284e-05,
|
| 1202 |
+
"loss": 0.8532,
|
| 1203 |
+
"step": 1670
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 0.033207487497776286,
|
| 1207 |
+
"grad_norm": 4.42426061630249,
|
| 1208 |
+
"learning_rate": 4.961030719295579e-05,
|
| 1209 |
+
"loss": 0.977,
|
| 1210 |
+
"step": 1680
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.03340515111383448,
|
| 1214 |
+
"grad_norm": 0.06116814911365509,
|
| 1215 |
+
"learning_rate": 4.960700191045329e-05,
|
| 1216 |
+
"loss": 0.5627,
|
| 1217 |
+
"step": 1690
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.03360281472989267,
|
| 1221 |
+
"grad_norm": 3.5464470386505127,
|
| 1222 |
+
"learning_rate": 4.960369662795079e-05,
|
| 1223 |
+
"loss": 0.6769,
|
| 1224 |
+
"step": 1700
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.03380047834595086,
|
| 1228 |
+
"grad_norm": 3.3844430446624756,
|
| 1229 |
+
"learning_rate": 4.96003913454483e-05,
|
| 1230 |
+
"loss": 1.448,
|
| 1231 |
+
"step": 1710
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.03399814196200905,
|
| 1235 |
+
"grad_norm": 3.592660427093506,
|
| 1236 |
+
"learning_rate": 4.9597086062945805e-05,
|
| 1237 |
+
"loss": 0.9505,
|
| 1238 |
+
"step": 1720
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.03419580557806724,
|
| 1242 |
+
"grad_norm": 3.355842113494873,
|
| 1243 |
+
"learning_rate": 4.959378078044331e-05,
|
| 1244 |
+
"loss": 1.0866,
|
| 1245 |
+
"step": 1730
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 0.034393469194125434,
|
| 1249 |
+
"grad_norm": 3.0316247940063477,
|
| 1250 |
+
"learning_rate": 4.959047549794081e-05,
|
| 1251 |
+
"loss": 0.6762,
|
| 1252 |
+
"step": 1740
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 0.03459113281018363,
|
| 1256 |
+
"grad_norm": 0.06703350692987442,
|
| 1257 |
+
"learning_rate": 4.9587170215438314e-05,
|
| 1258 |
+
"loss": 0.828,
|
| 1259 |
+
"step": 1750
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 0.03478879642624182,
|
| 1263 |
+
"grad_norm": 3.7815728187561035,
|
| 1264 |
+
"learning_rate": 4.9583864932935816e-05,
|
| 1265 |
+
"loss": 1.3015,
|
| 1266 |
+
"step": 1760
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 0.034986460042300015,
|
| 1270 |
+
"grad_norm": 3.3532164096832275,
|
| 1271 |
+
"learning_rate": 4.958055965043332e-05,
|
| 1272 |
+
"loss": 1.2062,
|
| 1273 |
+
"step": 1770
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.035184123658358206,
|
| 1277 |
+
"grad_norm": 3.1324400901794434,
|
| 1278 |
+
"learning_rate": 4.957725436793083e-05,
|
| 1279 |
+
"loss": 0.9484,
|
| 1280 |
+
"step": 1780
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.0353817872744164,
|
| 1284 |
+
"grad_norm": 0.03721948713064194,
|
| 1285 |
+
"learning_rate": 4.957394908542833e-05,
|
| 1286 |
+
"loss": 0.782,
|
| 1287 |
+
"step": 1790
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 0.03557945089047459,
|
| 1291 |
+
"grad_norm": 2.5899410247802734,
|
| 1292 |
+
"learning_rate": 4.9570643802925835e-05,
|
| 1293 |
+
"loss": 0.9005,
|
| 1294 |
+
"step": 1800
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 0.03577711450653278,
|
| 1298 |
+
"grad_norm": 0.14065319299697876,
|
| 1299 |
+
"learning_rate": 4.9567338520423344e-05,
|
| 1300 |
+
"loss": 1.2058,
|
| 1301 |
+
"step": 1810
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 0.03597477812259098,
|
| 1305 |
+
"grad_norm": 3.255124568939209,
|
| 1306 |
+
"learning_rate": 4.956403323792085e-05,
|
| 1307 |
+
"loss": 0.6861,
|
| 1308 |
+
"step": 1820
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 0.03617244173864917,
|
| 1312 |
+
"grad_norm": 2.9060869216918945,
|
| 1313 |
+
"learning_rate": 4.956072795541835e-05,
|
| 1314 |
+
"loss": 0.6472,
|
| 1315 |
+
"step": 1830
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 0.03637010535470736,
|
| 1319 |
+
"grad_norm": 0.10534641146659851,
|
| 1320 |
+
"learning_rate": 4.955742267291585e-05,
|
| 1321 |
+
"loss": 1.1794,
|
| 1322 |
+
"step": 1840
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 0.03656776897076555,
|
| 1326 |
+
"grad_norm": 3.3723835945129395,
|
| 1327 |
+
"learning_rate": 4.955411739041336e-05,
|
| 1328 |
+
"loss": 1.0936,
|
| 1329 |
+
"step": 1850
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.03676543258682374,
|
| 1333 |
+
"grad_norm": 3.3948323726654053,
|
| 1334 |
+
"learning_rate": 4.9550812107910865e-05,
|
| 1335 |
+
"loss": 1.0723,
|
| 1336 |
+
"step": 1860
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.036963096202881934,
|
| 1340 |
+
"grad_norm": 2.8475537300109863,
|
| 1341 |
+
"learning_rate": 4.954750682540837e-05,
|
| 1342 |
+
"loss": 1.1292,
|
| 1343 |
+
"step": 1870
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 0.037160759818940126,
|
| 1347 |
+
"grad_norm": 3.0720722675323486,
|
| 1348 |
+
"learning_rate": 4.954420154290588e-05,
|
| 1349 |
+
"loss": 0.4767,
|
| 1350 |
+
"step": 1880
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 0.03735842343499832,
|
| 1354 |
+
"grad_norm": 2.8738229274749756,
|
| 1355 |
+
"learning_rate": 4.954089626040338e-05,
|
| 1356 |
+
"loss": 0.8313,
|
| 1357 |
+
"step": 1890
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 0.037556087051056515,
|
| 1361 |
+
"grad_norm": 0.04877200722694397,
|
| 1362 |
+
"learning_rate": 4.9537590977900884e-05,
|
| 1363 |
+
"loss": 0.8162,
|
| 1364 |
+
"step": 1900
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 0.037753750667114706,
|
| 1368 |
+
"grad_norm": 0.15653006732463837,
|
| 1369 |
+
"learning_rate": 4.9534285695398386e-05,
|
| 1370 |
+
"loss": 1.117,
|
| 1371 |
+
"step": 1910
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 0.0379514142831729,
|
| 1375 |
+
"grad_norm": 0.09898870438337326,
|
| 1376 |
+
"learning_rate": 4.9530980412895896e-05,
|
| 1377 |
+
"loss": 0.8124,
|
| 1378 |
+
"step": 1920
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 0.03814907789923109,
|
| 1382 |
+
"grad_norm": 3.6042723655700684,
|
| 1383 |
+
"learning_rate": 4.95276751303934e-05,
|
| 1384 |
+
"loss": 0.978,
|
| 1385 |
+
"step": 1930
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 0.03834674151528928,
|
| 1389 |
+
"grad_norm": 2.490565776824951,
|
| 1390 |
+
"learning_rate": 4.95243698478909e-05,
|
| 1391 |
+
"loss": 0.4846,
|
| 1392 |
+
"step": 1940
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 0.03854440513134747,
|
| 1396 |
+
"grad_norm": 2.3419349193573,
|
| 1397 |
+
"learning_rate": 4.952106456538841e-05,
|
| 1398 |
+
"loss": 0.7827,
|
| 1399 |
+
"step": 1950
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 0.03874206874740566,
|
| 1403 |
+
"grad_norm": 2.327219009399414,
|
| 1404 |
+
"learning_rate": 4.9517759282885914e-05,
|
| 1405 |
+
"loss": 0.9868,
|
| 1406 |
+
"step": 1960
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 0.038939732363463854,
|
| 1410 |
+
"grad_norm": 0.18203556537628174,
|
| 1411 |
+
"learning_rate": 4.951445400038342e-05,
|
| 1412 |
+
"loss": 0.9514,
|
| 1413 |
+
"step": 1970
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 0.03913739597952205,
|
| 1417 |
+
"grad_norm": 2.613198757171631,
|
| 1418 |
+
"learning_rate": 4.951114871788092e-05,
|
| 1419 |
+
"loss": 0.7549,
|
| 1420 |
+
"step": 1980
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 0.03933505959558024,
|
| 1424 |
+
"grad_norm": 4.419915676116943,
|
| 1425 |
+
"learning_rate": 4.950784343537842e-05,
|
| 1426 |
+
"loss": 0.6902,
|
| 1427 |
+
"step": 1990
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 0.039532723211638435,
|
| 1431 |
+
"grad_norm": 0.04059309512376785,
|
| 1432 |
+
"learning_rate": 4.9504538152875926e-05,
|
| 1433 |
+
"loss": 0.8878,
|
| 1434 |
+
"step": 2000
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.039532723211638435,
|
| 1438 |
+
"eval_loss": 0.8785109519958496,
|
| 1439 |
+
"eval_runtime": 237.1008,
|
| 1440 |
+
"eval_samples_per_second": 89.852,
|
| 1441 |
+
"eval_steps_per_second": 11.232,
|
| 1442 |
+
"step": 2000
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.039730386827696626,
|
| 1446 |
+
"grad_norm": 2.7908315658569336,
|
| 1447 |
+
"learning_rate": 4.950123287037343e-05,
|
| 1448 |
+
"loss": 1.0453,
|
| 1449 |
+
"step": 2010
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.03992805044375482,
|
| 1453 |
+
"grad_norm": 0.19223953783512115,
|
| 1454 |
+
"learning_rate": 4.949792758787094e-05,
|
| 1455 |
+
"loss": 0.9602,
|
| 1456 |
+
"step": 2020
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 0.04012571405981301,
|
| 1460 |
+
"grad_norm": 2.7216477394104004,
|
| 1461 |
+
"learning_rate": 4.949462230536844e-05,
|
| 1462 |
+
"loss": 1.1187,
|
| 1463 |
+
"step": 2030
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 0.0403233776758712,
|
| 1467 |
+
"grad_norm": 3.306227922439575,
|
| 1468 |
+
"learning_rate": 4.9491317022865944e-05,
|
| 1469 |
+
"loss": 0.9302,
|
| 1470 |
+
"step": 2040
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 0.0405210412919294,
|
| 1474 |
+
"grad_norm": 3.259765386581421,
|
| 1475 |
+
"learning_rate": 4.948801174036345e-05,
|
| 1476 |
+
"loss": 0.9832,
|
| 1477 |
+
"step": 2050
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 0.04071870490798759,
|
| 1481 |
+
"grad_norm": 3.6598448753356934,
|
| 1482 |
+
"learning_rate": 4.9484706457860956e-05,
|
| 1483 |
+
"loss": 1.0918,
|
| 1484 |
+
"step": 2060
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 0.04091636852404578,
|
| 1488 |
+
"grad_norm": 3.6276862621307373,
|
| 1489 |
+
"learning_rate": 4.948140117535846e-05,
|
| 1490 |
+
"loss": 0.8682,
|
| 1491 |
+
"step": 2070
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 0.04111403214010397,
|
| 1495 |
+
"grad_norm": 3.547147035598755,
|
| 1496 |
+
"learning_rate": 4.947809589285596e-05,
|
| 1497 |
+
"loss": 0.5364,
|
| 1498 |
+
"step": 2080
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 0.04131169575616216,
|
| 1502 |
+
"grad_norm": 3.0729477405548096,
|
| 1503 |
+
"learning_rate": 4.947479061035347e-05,
|
| 1504 |
+
"loss": 1.2961,
|
| 1505 |
+
"step": 2090
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.041509359372220354,
|
| 1509 |
+
"grad_norm": 2.370206594467163,
|
| 1510 |
+
"learning_rate": 4.9471485327850975e-05,
|
| 1511 |
+
"loss": 0.6171,
|
| 1512 |
+
"step": 2100
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 0.041707022988278546,
|
| 1516 |
+
"grad_norm": 3.900785446166992,
|
| 1517 |
+
"learning_rate": 4.946818004534848e-05,
|
| 1518 |
+
"loss": 1.3057,
|
| 1519 |
+
"step": 2110
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 0.04190468660433674,
|
| 1523 |
+
"grad_norm": 3.226991653442383,
|
| 1524 |
+
"learning_rate": 4.946487476284598e-05,
|
| 1525 |
+
"loss": 1.0164,
|
| 1526 |
+
"step": 2120
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 0.042102350220394935,
|
| 1530 |
+
"grad_norm": 0.5991417169570923,
|
| 1531 |
+
"learning_rate": 4.946156948034349e-05,
|
| 1532 |
+
"loss": 1.4067,
|
| 1533 |
+
"step": 2130
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 0.042300013836453126,
|
| 1537 |
+
"grad_norm": 3.0779223442077637,
|
| 1538 |
+
"learning_rate": 4.945826419784099e-05,
|
| 1539 |
+
"loss": 0.7529,
|
| 1540 |
+
"step": 2140
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 0.04249767745251132,
|
| 1544 |
+
"grad_norm": 0.05838579684495926,
|
| 1545 |
+
"learning_rate": 4.9454958915338496e-05,
|
| 1546 |
+
"loss": 0.7606,
|
| 1547 |
+
"step": 2150
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 0.04269534106856951,
|
| 1551 |
+
"grad_norm": 3.432065486907959,
|
| 1552 |
+
"learning_rate": 4.9451653632836005e-05,
|
| 1553 |
+
"loss": 1.1331,
|
| 1554 |
+
"step": 2160
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 0.0428930046846277,
|
| 1558 |
+
"grad_norm": 2.9518392086029053,
|
| 1559 |
+
"learning_rate": 4.944834835033351e-05,
|
| 1560 |
+
"loss": 0.6276,
|
| 1561 |
+
"step": 2170
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 0.04309066830068589,
|
| 1565 |
+
"grad_norm": 0.06247523054480553,
|
| 1566 |
+
"learning_rate": 4.944504306783101e-05,
|
| 1567 |
+
"loss": 0.7537,
|
| 1568 |
+
"step": 2180
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 0.04328833191674408,
|
| 1572 |
+
"grad_norm": 2.4885005950927734,
|
| 1573 |
+
"learning_rate": 4.9441737785328514e-05,
|
| 1574 |
+
"loss": 1.1748,
|
| 1575 |
+
"step": 2190
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 0.043485995532802274,
|
| 1579 |
+
"grad_norm": 0.15998195111751556,
|
| 1580 |
+
"learning_rate": 4.9438432502826024e-05,
|
| 1581 |
+
"loss": 0.9554,
|
| 1582 |
+
"step": 2200
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 0.04368365914886047,
|
| 1586 |
+
"grad_norm": 3.214895248413086,
|
| 1587 |
+
"learning_rate": 4.9435127220323526e-05,
|
| 1588 |
+
"loss": 0.8503,
|
| 1589 |
+
"step": 2210
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 0.043881322764918664,
|
| 1593 |
+
"grad_norm": 0.03642280027270317,
|
| 1594 |
+
"learning_rate": 4.943182193782103e-05,
|
| 1595 |
+
"loss": 0.4855,
|
| 1596 |
+
"step": 2220
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 0.044078986380976855,
|
| 1600 |
+
"grad_norm": 2.707956075668335,
|
| 1601 |
+
"learning_rate": 4.942851665531853e-05,
|
| 1602 |
+
"loss": 1.0405,
|
| 1603 |
+
"step": 2230
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 0.044276649997035046,
|
| 1607 |
+
"grad_norm": 0.08170021325349808,
|
| 1608 |
+
"learning_rate": 4.9425211372816035e-05,
|
| 1609 |
+
"loss": 0.9093,
|
| 1610 |
+
"step": 2240
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 0.04447431361309324,
|
| 1614 |
+
"grad_norm": 2.322641611099243,
|
| 1615 |
+
"learning_rate": 4.942190609031354e-05,
|
| 1616 |
+
"loss": 1.357,
|
| 1617 |
+
"step": 2250
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 0.04467197722915143,
|
| 1621 |
+
"grad_norm": 2.5577518939971924,
|
| 1622 |
+
"learning_rate": 4.941860080781104e-05,
|
| 1623 |
+
"loss": 1.2809,
|
| 1624 |
+
"step": 2260
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 0.04486964084520962,
|
| 1628 |
+
"grad_norm": 2.471879243850708,
|
| 1629 |
+
"learning_rate": 4.941529552530855e-05,
|
| 1630 |
+
"loss": 1.1575,
|
| 1631 |
+
"step": 2270
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 0.04506730446126781,
|
| 1635 |
+
"grad_norm": 3.389317274093628,
|
| 1636 |
+
"learning_rate": 4.941199024280605e-05,
|
| 1637 |
+
"loss": 1.1552,
|
| 1638 |
+
"step": 2280
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 0.04526496807732601,
|
| 1642 |
+
"grad_norm": 0.11907307803630829,
|
| 1643 |
+
"learning_rate": 4.9408684960303556e-05,
|
| 1644 |
+
"loss": 0.685,
|
| 1645 |
+
"step": 2290
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 0.0454626316933842,
|
| 1649 |
+
"grad_norm": 3.5071091651916504,
|
| 1650 |
+
"learning_rate": 4.940537967780106e-05,
|
| 1651 |
+
"loss": 0.5446,
|
| 1652 |
+
"step": 2300
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 0.04566029530944239,
|
| 1656 |
+
"grad_norm": 2.466038227081299,
|
| 1657 |
+
"learning_rate": 4.940207439529857e-05,
|
| 1658 |
+
"loss": 0.9161,
|
| 1659 |
+
"step": 2310
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 0.04585795892550058,
|
| 1663 |
+
"grad_norm": 2.913646936416626,
|
| 1664 |
+
"learning_rate": 4.939876911279607e-05,
|
| 1665 |
+
"loss": 0.8943,
|
| 1666 |
+
"step": 2320
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 0.046055622541558774,
|
| 1670 |
+
"grad_norm": 2.976774215698242,
|
| 1671 |
+
"learning_rate": 4.9395463830293574e-05,
|
| 1672 |
+
"loss": 1.4517,
|
| 1673 |
+
"step": 2330
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 0.046253286157616966,
|
| 1677 |
+
"grad_norm": 0.20807339251041412,
|
| 1678 |
+
"learning_rate": 4.9392158547791084e-05,
|
| 1679 |
+
"loss": 0.535,
|
| 1680 |
+
"step": 2340
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 0.04645094977367516,
|
| 1684 |
+
"grad_norm": 3.2078752517700195,
|
| 1685 |
+
"learning_rate": 4.938885326528859e-05,
|
| 1686 |
+
"loss": 0.561,
|
| 1687 |
+
"step": 2350
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"epoch": 0.046648613389733355,
|
| 1691 |
+
"grad_norm": 2.7821130752563477,
|
| 1692 |
+
"learning_rate": 4.938554798278609e-05,
|
| 1693 |
+
"loss": 1.0918,
|
| 1694 |
+
"step": 2360
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 0.046846277005791546,
|
| 1698 |
+
"grad_norm": 0.18904615938663483,
|
| 1699 |
+
"learning_rate": 4.938224270028359e-05,
|
| 1700 |
+
"loss": 1.1504,
|
| 1701 |
+
"step": 2370
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 0.04704394062184974,
|
| 1705 |
+
"grad_norm": 2.9167234897613525,
|
| 1706 |
+
"learning_rate": 4.93789374177811e-05,
|
| 1707 |
+
"loss": 0.8307,
|
| 1708 |
+
"step": 2380
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 0.04724160423790793,
|
| 1712 |
+
"grad_norm": 2.7260210514068604,
|
| 1713 |
+
"learning_rate": 4.9375632135278605e-05,
|
| 1714 |
+
"loss": 0.7618,
|
| 1715 |
+
"step": 2390
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 0.04743926785396612,
|
| 1719 |
+
"grad_norm": 2.3594138622283936,
|
| 1720 |
+
"learning_rate": 4.937232685277611e-05,
|
| 1721 |
+
"loss": 0.4669,
|
| 1722 |
+
"step": 2400
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 0.04763693147002431,
|
| 1726 |
+
"grad_norm": 2.587303400039673,
|
| 1727 |
+
"learning_rate": 4.936902157027362e-05,
|
| 1728 |
+
"loss": 0.6653,
|
| 1729 |
+
"step": 2410
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 0.0478345950860825,
|
| 1733 |
+
"grad_norm": 0.07142865657806396,
|
| 1734 |
+
"learning_rate": 4.936571628777112e-05,
|
| 1735 |
+
"loss": 0.8185,
|
| 1736 |
+
"step": 2420
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 0.048032258702140694,
|
| 1740 |
+
"grad_norm": 3.059152603149414,
|
| 1741 |
+
"learning_rate": 4.936241100526862e-05,
|
| 1742 |
+
"loss": 0.9088,
|
| 1743 |
+
"step": 2430
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 0.04822992231819889,
|
| 1747 |
+
"grad_norm": 0.1070665493607521,
|
| 1748 |
+
"learning_rate": 4.9359105722766126e-05,
|
| 1749 |
+
"loss": 1.1149,
|
| 1750 |
+
"step": 2440
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 0.048427585934257084,
|
| 1754 |
+
"grad_norm": 2.317112684249878,
|
| 1755 |
+
"learning_rate": 4.9355800440263636e-05,
|
| 1756 |
+
"loss": 0.5108,
|
| 1757 |
+
"step": 2450
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 0.048625249550315275,
|
| 1761 |
+
"grad_norm": 2.8447506427764893,
|
| 1762 |
+
"learning_rate": 4.935249515776114e-05,
|
| 1763 |
+
"loss": 1.0359,
|
| 1764 |
+
"step": 2460
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 0.048822913166373466,
|
| 1768 |
+
"grad_norm": 3.0613043308258057,
|
| 1769 |
+
"learning_rate": 4.934918987525864e-05,
|
| 1770 |
+
"loss": 0.7497,
|
| 1771 |
+
"step": 2470
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 0.04902057678243166,
|
| 1775 |
+
"grad_norm": 2.9260942935943604,
|
| 1776 |
+
"learning_rate": 4.9345884592756144e-05,
|
| 1777 |
+
"loss": 0.9751,
|
| 1778 |
+
"step": 2480
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 0.04921824039848985,
|
| 1782 |
+
"grad_norm": 0.17592434585094452,
|
| 1783 |
+
"learning_rate": 4.934257931025365e-05,
|
| 1784 |
+
"loss": 0.9219,
|
| 1785 |
+
"step": 2490
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 0.04941590401454804,
|
| 1789 |
+
"grad_norm": 0.07734190672636032,
|
| 1790 |
+
"learning_rate": 4.933927402775116e-05,
|
| 1791 |
+
"loss": 0.2931,
|
| 1792 |
+
"step": 2500
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 0.04941590401454804,
|
| 1796 |
+
"eval_loss": 0.8793515563011169,
|
| 1797 |
+
"eval_runtime": 237.1956,
|
| 1798 |
+
"eval_samples_per_second": 89.816,
|
| 1799 |
+
"eval_steps_per_second": 11.227,
|
| 1800 |
+
"step": 2500
|
| 1801 |
+
}
|
| 1802 |
+
],
|
| 1803 |
+
"logging_steps": 10,
|
| 1804 |
+
"max_steps": 151773,
|
| 1805 |
+
"num_input_tokens_seen": 0,
|
| 1806 |
+
"num_train_epochs": 3,
|
| 1807 |
+
"save_steps": 500,
|
| 1808 |
+
"stateful_callbacks": {
|
| 1809 |
+
"TrainerControl": {
|
| 1810 |
+
"args": {
|
| 1811 |
+
"should_epoch_stop": false,
|
| 1812 |
+
"should_evaluate": false,
|
| 1813 |
+
"should_log": false,
|
| 1814 |
+
"should_save": true,
|
| 1815 |
+
"should_training_stop": false
|
| 1816 |
+
},
|
| 1817 |
+
"attributes": {}
|
| 1818 |
+
}
|
| 1819 |
+
},
|
| 1820 |
+
"total_flos": 1.045168128e+16,
|
| 1821 |
+
"train_batch_size": 8,
|
| 1822 |
+
"trial_name": null,
|
| 1823 |
+
"trial_params": null
|
| 1824 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5852d5e962258e5f74ecbf70ae63c57ef8107ae9b7100b4eba658baf97dadba7
|
| 3 |
+
size 5713
|