arabic lang model v1 18000 ,0.76
Browse files- config.json +43 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1294 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 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 |
+
"reorder_and_upcast_attn": false,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 5,
|
| 35 |
+
"repetition_penalty": 3.0,
|
| 36 |
+
"top_p": 0.95
|
| 37 |
+
}
|
| 38 |
+
},
|
| 39 |
+
"torch_dtype": "float32",
|
| 40 |
+
"transformers_version": "4.55.0",
|
| 41 |
+
"use_cache": true,
|
| 42 |
+
"vocab_size": 64000
|
| 43 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.55.0"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd53bd09a142232028fdb15f68d0868f685361208fc3cbf986245f9116b3401b
|
| 3 |
+
size 539992704
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c51b3524735a0d9684d03edb725ee3e1e828e67d9a4d10b323e7299a833ebc97
|
| 3 |
+
size 1080079290
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb52cd183a737416b549c539d549c5f95c079ec4e1802079241e54bd663cd5a0
|
| 3 |
+
size 14244
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c59b6372878a323eca88cc61bfa524dbdc722a5b1c902691464791b521616f80
|
| 3 |
+
size 988
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:422e363f834b383564c224b14c73061c3ab13f55b6c3887dbc1c420c3e2e0585
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1294 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.3910355486862442,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 18000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0077279752704791345,
|
| 14 |
+
"grad_norm": 2.382732391357422,
|
| 15 |
+
"learning_rate": 4.987764039155075e-05,
|
| 16 |
+
"loss": 2.2319,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.015455950540958269,
|
| 21 |
+
"grad_norm": 1.1049453020095825,
|
| 22 |
+
"learning_rate": 4.974884080370943e-05,
|
| 23 |
+
"loss": 1.4633,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.023183925811437404,
|
| 28 |
+
"grad_norm": 2.012437582015991,
|
| 29 |
+
"learning_rate": 4.962004121586811e-05,
|
| 30 |
+
"loss": 1.4898,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.030911901081916538,
|
| 35 |
+
"grad_norm": 2.1583738327026367,
|
| 36 |
+
"learning_rate": 4.949124162802679e-05,
|
| 37 |
+
"loss": 1.2682,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03863987635239567,
|
| 42 |
+
"grad_norm": 1.3249982595443726,
|
| 43 |
+
"learning_rate": 4.9362442040185474e-05,
|
| 44 |
+
"loss": 1.3189,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04636785162287481,
|
| 49 |
+
"grad_norm": 1.0867700576782227,
|
| 50 |
+
"learning_rate": 4.923364245234415e-05,
|
| 51 |
+
"loss": 1.2482,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.05409582689335394,
|
| 56 |
+
"grad_norm": 0.9986234307289124,
|
| 57 |
+
"learning_rate": 4.910484286450284e-05,
|
| 58 |
+
"loss": 1.1299,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.061823802163833076,
|
| 63 |
+
"grad_norm": 0.8884647488594055,
|
| 64 |
+
"learning_rate": 4.897604327666152e-05,
|
| 65 |
+
"loss": 1.2575,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0695517774343122,
|
| 70 |
+
"grad_norm": 1.6838749647140503,
|
| 71 |
+
"learning_rate": 4.8847243688820196e-05,
|
| 72 |
+
"loss": 1.1865,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.07727975270479134,
|
| 77 |
+
"grad_norm": 2.3805487155914307,
|
| 78 |
+
"learning_rate": 4.8718444100978875e-05,
|
| 79 |
+
"loss": 1.1084,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08500772797527048,
|
| 84 |
+
"grad_norm": 1.653010606765747,
|
| 85 |
+
"learning_rate": 4.858964451313756e-05,
|
| 86 |
+
"loss": 1.1401,
|
| 87 |
+
"step": 1100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.09273570324574962,
|
| 91 |
+
"grad_norm": 1.6932406425476074,
|
| 92 |
+
"learning_rate": 4.846084492529624e-05,
|
| 93 |
+
"loss": 1.2395,
|
| 94 |
+
"step": 1200
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10046367851622875,
|
| 98 |
+
"grad_norm": 1.2242937088012695,
|
| 99 |
+
"learning_rate": 4.8332045337454925e-05,
|
| 100 |
+
"loss": 1.1328,
|
| 101 |
+
"step": 1300
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.10819165378670788,
|
| 105 |
+
"grad_norm": 2.8659744262695312,
|
| 106 |
+
"learning_rate": 4.8203245749613604e-05,
|
| 107 |
+
"loss": 1.1238,
|
| 108 |
+
"step": 1400
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11591962905718702,
|
| 112 |
+
"grad_norm": 2.079432249069214,
|
| 113 |
+
"learning_rate": 4.807444616177228e-05,
|
| 114 |
+
"loss": 1.0579,
|
| 115 |
+
"step": 1500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.12364760432766615,
|
| 119 |
+
"grad_norm": 0.5808582901954651,
|
| 120 |
+
"learning_rate": 4.794564657393096e-05,
|
| 121 |
+
"loss": 0.9646,
|
| 122 |
+
"step": 1600
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.13137557959814528,
|
| 126 |
+
"grad_norm": 1.380146861076355,
|
| 127 |
+
"learning_rate": 4.781684698608965e-05,
|
| 128 |
+
"loss": 1.1086,
|
| 129 |
+
"step": 1700
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1391035548686244,
|
| 133 |
+
"grad_norm": 1.4197566509246826,
|
| 134 |
+
"learning_rate": 4.7688047398248325e-05,
|
| 135 |
+
"loss": 1.1619,
|
| 136 |
+
"step": 1800
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.14683153013910355,
|
| 140 |
+
"grad_norm": 1.1287027597427368,
|
| 141 |
+
"learning_rate": 4.755924781040701e-05,
|
| 142 |
+
"loss": 1.0074,
|
| 143 |
+
"step": 1900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1545595054095827,
|
| 147 |
+
"grad_norm": 1.1685930490493774,
|
| 148 |
+
"learning_rate": 4.743044822256569e-05,
|
| 149 |
+
"loss": 1.1573,
|
| 150 |
+
"step": 2000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.16228748068006182,
|
| 154 |
+
"grad_norm": 1.008876919746399,
|
| 155 |
+
"learning_rate": 4.730164863472437e-05,
|
| 156 |
+
"loss": 1.0858,
|
| 157 |
+
"step": 2100
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.17001545595054096,
|
| 161 |
+
"grad_norm": 0.9628033638000488,
|
| 162 |
+
"learning_rate": 4.717284904688305e-05,
|
| 163 |
+
"loss": 1.1267,
|
| 164 |
+
"step": 2200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.1777434312210201,
|
| 168 |
+
"grad_norm": 1.298677682876587,
|
| 169 |
+
"learning_rate": 4.704404945904173e-05,
|
| 170 |
+
"loss": 1.0804,
|
| 171 |
+
"step": 2300
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.18547140649149924,
|
| 175 |
+
"grad_norm": 1.6227502822875977,
|
| 176 |
+
"learning_rate": 4.691524987120041e-05,
|
| 177 |
+
"loss": 1.1371,
|
| 178 |
+
"step": 2400
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.19319938176197837,
|
| 182 |
+
"grad_norm": 1.6179583072662354,
|
| 183 |
+
"learning_rate": 4.67864502833591e-05,
|
| 184 |
+
"loss": 1.0538,
|
| 185 |
+
"step": 2500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2009273570324575,
|
| 189 |
+
"grad_norm": 1.242937445640564,
|
| 190 |
+
"learning_rate": 4.6657650695517776e-05,
|
| 191 |
+
"loss": 0.9714,
|
| 192 |
+
"step": 2600
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.20865533230293662,
|
| 196 |
+
"grad_norm": 1.4579142332077026,
|
| 197 |
+
"learning_rate": 4.6528851107676455e-05,
|
| 198 |
+
"loss": 1.1136,
|
| 199 |
+
"step": 2700
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.21638330757341576,
|
| 203 |
+
"grad_norm": 1.5873304605484009,
|
| 204 |
+
"learning_rate": 4.6400051519835134e-05,
|
| 205 |
+
"loss": 1.0059,
|
| 206 |
+
"step": 2800
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2241112828438949,
|
| 210 |
+
"grad_norm": 1.6348575353622437,
|
| 211 |
+
"learning_rate": 4.627125193199382e-05,
|
| 212 |
+
"loss": 1.009,
|
| 213 |
+
"step": 2900
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.23183925811437403,
|
| 217 |
+
"grad_norm": 1.373192548751831,
|
| 218 |
+
"learning_rate": 4.6142452344152505e-05,
|
| 219 |
+
"loss": 1.0069,
|
| 220 |
+
"step": 3000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.23956723338485317,
|
| 224 |
+
"grad_norm": 0.9459308385848999,
|
| 225 |
+
"learning_rate": 4.6013652756311184e-05,
|
| 226 |
+
"loss": 0.9666,
|
| 227 |
+
"step": 3100
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2472952086553323,
|
| 231 |
+
"grad_norm": 1.5119107961654663,
|
| 232 |
+
"learning_rate": 4.588485316846986e-05,
|
| 233 |
+
"loss": 1.0447,
|
| 234 |
+
"step": 3200
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2550231839258114,
|
| 238 |
+
"grad_norm": 1.3911887407302856,
|
| 239 |
+
"learning_rate": 4.575605358062854e-05,
|
| 240 |
+
"loss": 1.0636,
|
| 241 |
+
"step": 3300
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.26275115919629055,
|
| 245 |
+
"grad_norm": 1.434175729751587,
|
| 246 |
+
"learning_rate": 4.562725399278723e-05,
|
| 247 |
+
"loss": 1.0425,
|
| 248 |
+
"step": 3400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2704791344667697,
|
| 252 |
+
"grad_norm": 1.7139261960983276,
|
| 253 |
+
"learning_rate": 4.5498454404945905e-05,
|
| 254 |
+
"loss": 1.0795,
|
| 255 |
+
"step": 3500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.2782071097372488,
|
| 259 |
+
"grad_norm": 0.8860535025596619,
|
| 260 |
+
"learning_rate": 4.536965481710459e-05,
|
| 261 |
+
"loss": 0.9276,
|
| 262 |
+
"step": 3600
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.28593508500772796,
|
| 266 |
+
"grad_norm": 1.6701558828353882,
|
| 267 |
+
"learning_rate": 4.524085522926327e-05,
|
| 268 |
+
"loss": 0.9323,
|
| 269 |
+
"step": 3700
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.2936630602782071,
|
| 273 |
+
"grad_norm": 1.3985803127288818,
|
| 274 |
+
"learning_rate": 4.511205564142195e-05,
|
| 275 |
+
"loss": 0.9605,
|
| 276 |
+
"step": 3800
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.30139103554868624,
|
| 280 |
+
"grad_norm": 0.7279998064041138,
|
| 281 |
+
"learning_rate": 4.498325605358063e-05,
|
| 282 |
+
"loss": 0.9523,
|
| 283 |
+
"step": 3900
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3091190108191654,
|
| 287 |
+
"grad_norm": 0.7744489312171936,
|
| 288 |
+
"learning_rate": 4.485445646573931e-05,
|
| 289 |
+
"loss": 1.03,
|
| 290 |
+
"step": 4000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3168469860896445,
|
| 294 |
+
"grad_norm": 1.330718755722046,
|
| 295 |
+
"learning_rate": 4.472565687789799e-05,
|
| 296 |
+
"loss": 1.0305,
|
| 297 |
+
"step": 4100
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.32457496136012365,
|
| 301 |
+
"grad_norm": 1.212430477142334,
|
| 302 |
+
"learning_rate": 4.459685729005668e-05,
|
| 303 |
+
"loss": 1.055,
|
| 304 |
+
"step": 4200
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3323029366306028,
|
| 308 |
+
"grad_norm": 1.0525803565979004,
|
| 309 |
+
"learning_rate": 4.4468057702215356e-05,
|
| 310 |
+
"loss": 1.0114,
|
| 311 |
+
"step": 4300
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3400309119010819,
|
| 315 |
+
"grad_norm": 1.147479772567749,
|
| 316 |
+
"learning_rate": 4.433925811437404e-05,
|
| 317 |
+
"loss": 0.9928,
|
| 318 |
+
"step": 4400
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.34775888717156106,
|
| 322 |
+
"grad_norm": 1.313454508781433,
|
| 323 |
+
"learning_rate": 4.4210458526532714e-05,
|
| 324 |
+
"loss": 0.974,
|
| 325 |
+
"step": 4500
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3554868624420402,
|
| 329 |
+
"grad_norm": 1.4218354225158691,
|
| 330 |
+
"learning_rate": 4.40816589386914e-05,
|
| 331 |
+
"loss": 0.9413,
|
| 332 |
+
"step": 4600
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.36321483771251933,
|
| 336 |
+
"grad_norm": 1.2902942895889282,
|
| 337 |
+
"learning_rate": 4.395285935085008e-05,
|
| 338 |
+
"loss": 1.002,
|
| 339 |
+
"step": 4700
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.37094281298299847,
|
| 343 |
+
"grad_norm": 0.638493001461029,
|
| 344 |
+
"learning_rate": 4.3824059763008763e-05,
|
| 345 |
+
"loss": 0.9175,
|
| 346 |
+
"step": 4800
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3786707882534776,
|
| 350 |
+
"grad_norm": 1.4392635822296143,
|
| 351 |
+
"learning_rate": 4.369526017516744e-05,
|
| 352 |
+
"loss": 0.89,
|
| 353 |
+
"step": 4900
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.38639876352395675,
|
| 357 |
+
"grad_norm": 2.024583339691162,
|
| 358 |
+
"learning_rate": 4.356646058732613e-05,
|
| 359 |
+
"loss": 0.9731,
|
| 360 |
+
"step": 5000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.3941267387944359,
|
| 364 |
+
"grad_norm": 1.8554579019546509,
|
| 365 |
+
"learning_rate": 4.34376609994848e-05,
|
| 366 |
+
"loss": 0.9251,
|
| 367 |
+
"step": 5100
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.401854714064915,
|
| 371 |
+
"grad_norm": 0.7937321662902832,
|
| 372 |
+
"learning_rate": 4.3308861411643485e-05,
|
| 373 |
+
"loss": 0.9758,
|
| 374 |
+
"step": 5200
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.4095826893353941,
|
| 378 |
+
"grad_norm": 1.3737086057662964,
|
| 379 |
+
"learning_rate": 4.3180061823802164e-05,
|
| 380 |
+
"loss": 1.0431,
|
| 381 |
+
"step": 5300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.41731066460587324,
|
| 385 |
+
"grad_norm": 1.1854668855667114,
|
| 386 |
+
"learning_rate": 4.305126223596085e-05,
|
| 387 |
+
"loss": 0.9832,
|
| 388 |
+
"step": 5400
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4250386398763524,
|
| 392 |
+
"grad_norm": 1.4984533786773682,
|
| 393 |
+
"learning_rate": 4.292246264811953e-05,
|
| 394 |
+
"loss": 0.9949,
|
| 395 |
+
"step": 5500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4327666151468315,
|
| 399 |
+
"grad_norm": 0.8790501356124878,
|
| 400 |
+
"learning_rate": 4.2793663060278214e-05,
|
| 401 |
+
"loss": 1.0039,
|
| 402 |
+
"step": 5600
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.44049459041731065,
|
| 406 |
+
"grad_norm": 0.7149144411087036,
|
| 407 |
+
"learning_rate": 4.2664863472436886e-05,
|
| 408 |
+
"loss": 0.9687,
|
| 409 |
+
"step": 5700
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4482225656877898,
|
| 413 |
+
"grad_norm": 1.146522045135498,
|
| 414 |
+
"learning_rate": 4.253606388459557e-05,
|
| 415 |
+
"loss": 0.9247,
|
| 416 |
+
"step": 5800
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.4559505409582689,
|
| 420 |
+
"grad_norm": 1.8279156684875488,
|
| 421 |
+
"learning_rate": 4.240726429675425e-05,
|
| 422 |
+
"loss": 0.998,
|
| 423 |
+
"step": 5900
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.46367851622874806,
|
| 427 |
+
"grad_norm": 0.974434494972229,
|
| 428 |
+
"learning_rate": 4.2278464708912936e-05,
|
| 429 |
+
"loss": 1.0235,
|
| 430 |
+
"step": 6000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.4714064914992272,
|
| 434 |
+
"grad_norm": 1.2241114377975464,
|
| 435 |
+
"learning_rate": 4.2149665121071615e-05,
|
| 436 |
+
"loss": 0.854,
|
| 437 |
+
"step": 6100
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.47913446676970634,
|
| 441 |
+
"grad_norm": 1.2750343084335327,
|
| 442 |
+
"learning_rate": 4.20208655332303e-05,
|
| 443 |
+
"loss": 1.005,
|
| 444 |
+
"step": 6200
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.4868624420401855,
|
| 448 |
+
"grad_norm": 1.21708345413208,
|
| 449 |
+
"learning_rate": 4.189206594538897e-05,
|
| 450 |
+
"loss": 0.9381,
|
| 451 |
+
"step": 6300
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.4945904173106646,
|
| 455 |
+
"grad_norm": 1.3617980480194092,
|
| 456 |
+
"learning_rate": 4.176326635754766e-05,
|
| 457 |
+
"loss": 0.8568,
|
| 458 |
+
"step": 6400
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5023183925811437,
|
| 462 |
+
"grad_norm": 1.8272175788879395,
|
| 463 |
+
"learning_rate": 4.163446676970634e-05,
|
| 464 |
+
"loss": 0.8578,
|
| 465 |
+
"step": 6500
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5100463678516228,
|
| 469 |
+
"grad_norm": 0.9534468650817871,
|
| 470 |
+
"learning_rate": 4.150566718186502e-05,
|
| 471 |
+
"loss": 0.9316,
|
| 472 |
+
"step": 6600
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.517774343122102,
|
| 476 |
+
"grad_norm": 0.8899122476577759,
|
| 477 |
+
"learning_rate": 4.13768675940237e-05,
|
| 478 |
+
"loss": 0.8945,
|
| 479 |
+
"step": 6700
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5255023183925811,
|
| 483 |
+
"grad_norm": 0.9442152380943298,
|
| 484 |
+
"learning_rate": 4.1248068006182387e-05,
|
| 485 |
+
"loss": 0.8847,
|
| 486 |
+
"step": 6800
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5332302936630603,
|
| 490 |
+
"grad_norm": 1.1126829385757446,
|
| 491 |
+
"learning_rate": 4.111926841834106e-05,
|
| 492 |
+
"loss": 0.9536,
|
| 493 |
+
"step": 6900
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5409582689335394,
|
| 497 |
+
"grad_norm": 1.2574111223220825,
|
| 498 |
+
"learning_rate": 4.0990468830499744e-05,
|
| 499 |
+
"loss": 0.9258,
|
| 500 |
+
"step": 7000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5486862442040186,
|
| 504 |
+
"grad_norm": 1.5313997268676758,
|
| 505 |
+
"learning_rate": 4.086166924265842e-05,
|
| 506 |
+
"loss": 0.914,
|
| 507 |
+
"step": 7100
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.5564142194744977,
|
| 511 |
+
"grad_norm": 1.064965009689331,
|
| 512 |
+
"learning_rate": 4.073286965481711e-05,
|
| 513 |
+
"loss": 0.9365,
|
| 514 |
+
"step": 7200
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5641421947449768,
|
| 518 |
+
"grad_norm": 1.4868042469024658,
|
| 519 |
+
"learning_rate": 4.060407006697579e-05,
|
| 520 |
+
"loss": 0.9697,
|
| 521 |
+
"step": 7300
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.5718701700154559,
|
| 525 |
+
"grad_norm": 0.8288936018943787,
|
| 526 |
+
"learning_rate": 4.047527047913447e-05,
|
| 527 |
+
"loss": 0.8882,
|
| 528 |
+
"step": 7400
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.5795981452859351,
|
| 532 |
+
"grad_norm": 0.5883311033248901,
|
| 533 |
+
"learning_rate": 4.0346470891293145e-05,
|
| 534 |
+
"loss": 0.9866,
|
| 535 |
+
"step": 7500
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.5873261205564142,
|
| 539 |
+
"grad_norm": 0.6529534459114075,
|
| 540 |
+
"learning_rate": 4.021767130345183e-05,
|
| 541 |
+
"loss": 0.9215,
|
| 542 |
+
"step": 7600
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.5950540958268934,
|
| 546 |
+
"grad_norm": 0.7607959508895874,
|
| 547 |
+
"learning_rate": 4.008887171561051e-05,
|
| 548 |
+
"loss": 0.9043,
|
| 549 |
+
"step": 7700
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6027820710973725,
|
| 553 |
+
"grad_norm": 1.3604129552841187,
|
| 554 |
+
"learning_rate": 3.9960072127769195e-05,
|
| 555 |
+
"loss": 0.9164,
|
| 556 |
+
"step": 7800
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6105100463678517,
|
| 560 |
+
"grad_norm": 1.1792867183685303,
|
| 561 |
+
"learning_rate": 3.9831272539927873e-05,
|
| 562 |
+
"loss": 0.9893,
|
| 563 |
+
"step": 7900
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6182380216383307,
|
| 567 |
+
"grad_norm": 0.9397594332695007,
|
| 568 |
+
"learning_rate": 3.970247295208656e-05,
|
| 569 |
+
"loss": 0.9113,
|
| 570 |
+
"step": 8000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6259659969088099,
|
| 574 |
+
"grad_norm": 0.5270503759384155,
|
| 575 |
+
"learning_rate": 3.957367336424523e-05,
|
| 576 |
+
"loss": 0.9062,
|
| 577 |
+
"step": 8100
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.633693972179289,
|
| 581 |
+
"grad_norm": 0.9222223162651062,
|
| 582 |
+
"learning_rate": 3.944487377640392e-05,
|
| 583 |
+
"loss": 0.9085,
|
| 584 |
+
"step": 8200
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6414219474497682,
|
| 588 |
+
"grad_norm": 1.1142535209655762,
|
| 589 |
+
"learning_rate": 3.9316074188562595e-05,
|
| 590 |
+
"loss": 0.9342,
|
| 591 |
+
"step": 8300
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.6491499227202473,
|
| 595 |
+
"grad_norm": 0.9540684223175049,
|
| 596 |
+
"learning_rate": 3.918727460072128e-05,
|
| 597 |
+
"loss": 0.9318,
|
| 598 |
+
"step": 8400
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6568778979907264,
|
| 602 |
+
"grad_norm": 1.5218793153762817,
|
| 603 |
+
"learning_rate": 3.905847501287996e-05,
|
| 604 |
+
"loss": 0.952,
|
| 605 |
+
"step": 8500
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.6646058732612056,
|
| 609 |
+
"grad_norm": 0.6363744735717773,
|
| 610 |
+
"learning_rate": 3.8929675425038645e-05,
|
| 611 |
+
"loss": 0.8996,
|
| 612 |
+
"step": 8600
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.6723338485316847,
|
| 616 |
+
"grad_norm": 1.2216317653656006,
|
| 617 |
+
"learning_rate": 3.880087583719732e-05,
|
| 618 |
+
"loss": 0.9585,
|
| 619 |
+
"step": 8700
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.6800618238021638,
|
| 623 |
+
"grad_norm": 0.5407654643058777,
|
| 624 |
+
"learning_rate": 3.8672076249356e-05,
|
| 625 |
+
"loss": 0.9532,
|
| 626 |
+
"step": 8800
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.6877897990726429,
|
| 630 |
+
"grad_norm": 0.936690092086792,
|
| 631 |
+
"learning_rate": 3.854327666151468e-05,
|
| 632 |
+
"loss": 0.8751,
|
| 633 |
+
"step": 8900
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.6955177743431221,
|
| 637 |
+
"grad_norm": 1.137471079826355,
|
| 638 |
+
"learning_rate": 3.841447707367337e-05,
|
| 639 |
+
"loss": 0.8881,
|
| 640 |
+
"step": 9000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7032457496136012,
|
| 644 |
+
"grad_norm": 0.28785666823387146,
|
| 645 |
+
"learning_rate": 3.828567748583205e-05,
|
| 646 |
+
"loss": 0.9673,
|
| 647 |
+
"step": 9100
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7109737248840804,
|
| 651 |
+
"grad_norm": 0.8596464991569519,
|
| 652 |
+
"learning_rate": 3.815687789799073e-05,
|
| 653 |
+
"loss": 0.9919,
|
| 654 |
+
"step": 9200
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7187017001545595,
|
| 658 |
+
"grad_norm": 1.2839220762252808,
|
| 659 |
+
"learning_rate": 3.802807831014941e-05,
|
| 660 |
+
"loss": 0.9603,
|
| 661 |
+
"step": 9300
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7264296754250387,
|
| 665 |
+
"grad_norm": 1.0451067686080933,
|
| 666 |
+
"learning_rate": 3.789927872230809e-05,
|
| 667 |
+
"loss": 0.9291,
|
| 668 |
+
"step": 9400
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7341576506955177,
|
| 672 |
+
"grad_norm": 0.7771898508071899,
|
| 673 |
+
"learning_rate": 3.7770479134466775e-05,
|
| 674 |
+
"loss": 0.9414,
|
| 675 |
+
"step": 9500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.7418856259659969,
|
| 679 |
+
"grad_norm": 0.885811984539032,
|
| 680 |
+
"learning_rate": 3.7641679546625453e-05,
|
| 681 |
+
"loss": 0.8987,
|
| 682 |
+
"step": 9600
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.749613601236476,
|
| 686 |
+
"grad_norm": 1.3268210887908936,
|
| 687 |
+
"learning_rate": 3.751287995878414e-05,
|
| 688 |
+
"loss": 0.9359,
|
| 689 |
+
"step": 9700
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7573415765069552,
|
| 693 |
+
"grad_norm": 0.6823452711105347,
|
| 694 |
+
"learning_rate": 3.738408037094282e-05,
|
| 695 |
+
"loss": 0.8704,
|
| 696 |
+
"step": 9800
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7650695517774343,
|
| 700 |
+
"grad_norm": 0.8156200647354126,
|
| 701 |
+
"learning_rate": 3.7255280783101497e-05,
|
| 702 |
+
"loss": 0.8954,
|
| 703 |
+
"step": 9900
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.7727975270479135,
|
| 707 |
+
"grad_norm": 0.8661110997200012,
|
| 708 |
+
"learning_rate": 3.7126481195260175e-05,
|
| 709 |
+
"loss": 0.8889,
|
| 710 |
+
"step": 10000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.7805255023183926,
|
| 714 |
+
"grad_norm": 0.5882070064544678,
|
| 715 |
+
"learning_rate": 3.699768160741886e-05,
|
| 716 |
+
"loss": 0.916,
|
| 717 |
+
"step": 10100
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.7882534775888718,
|
| 721 |
+
"grad_norm": 0.9691454768180847,
|
| 722 |
+
"learning_rate": 3.686888201957754e-05,
|
| 723 |
+
"loss": 0.834,
|
| 724 |
+
"step": 10200
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.7959814528593508,
|
| 728 |
+
"grad_norm": 1.3009424209594727,
|
| 729 |
+
"learning_rate": 3.6740082431736225e-05,
|
| 730 |
+
"loss": 0.9496,
|
| 731 |
+
"step": 10300
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.80370942812983,
|
| 735 |
+
"grad_norm": 1.1349143981933594,
|
| 736 |
+
"learning_rate": 3.6611282843894904e-05,
|
| 737 |
+
"loss": 0.8839,
|
| 738 |
+
"step": 10400
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8114374034003091,
|
| 742 |
+
"grad_norm": 0.924774706363678,
|
| 743 |
+
"learning_rate": 3.648248325605358e-05,
|
| 744 |
+
"loss": 0.9256,
|
| 745 |
+
"step": 10500
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8191653786707882,
|
| 749 |
+
"grad_norm": 0.7730056047439575,
|
| 750 |
+
"learning_rate": 3.635368366821226e-05,
|
| 751 |
+
"loss": 0.9233,
|
| 752 |
+
"step": 10600
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8268933539412674,
|
| 756 |
+
"grad_norm": 1.2240443229675293,
|
| 757 |
+
"learning_rate": 3.622488408037095e-05,
|
| 758 |
+
"loss": 0.9311,
|
| 759 |
+
"step": 10700
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.8346213292117465,
|
| 763 |
+
"grad_norm": 1.022357702255249,
|
| 764 |
+
"learning_rate": 3.6096084492529626e-05,
|
| 765 |
+
"loss": 0.8764,
|
| 766 |
+
"step": 10800
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.8423493044822257,
|
| 770 |
+
"grad_norm": 0.9591399431228638,
|
| 771 |
+
"learning_rate": 3.596728490468831e-05,
|
| 772 |
+
"loss": 0.8918,
|
| 773 |
+
"step": 10900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.8500772797527048,
|
| 777 |
+
"grad_norm": 0.5928884744644165,
|
| 778 |
+
"learning_rate": 3.583848531684699e-05,
|
| 779 |
+
"loss": 0.9446,
|
| 780 |
+
"step": 11000
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.8578052550231839,
|
| 784 |
+
"grad_norm": 1.6488679647445679,
|
| 785 |
+
"learning_rate": 3.5710973724884085e-05,
|
| 786 |
+
"loss": 0.8841,
|
| 787 |
+
"step": 11100
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.865533230293663,
|
| 791 |
+
"grad_norm": 1.309877872467041,
|
| 792 |
+
"learning_rate": 3.5582174137042763e-05,
|
| 793 |
+
"loss": 0.849,
|
| 794 |
+
"step": 11200
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.8732612055641422,
|
| 798 |
+
"grad_norm": 1.1240078210830688,
|
| 799 |
+
"learning_rate": 3.545337454920145e-05,
|
| 800 |
+
"loss": 0.9454,
|
| 801 |
+
"step": 11300
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.8809891808346213,
|
| 805 |
+
"grad_norm": 0.9316561222076416,
|
| 806 |
+
"learning_rate": 3.532457496136012e-05,
|
| 807 |
+
"loss": 0.9605,
|
| 808 |
+
"step": 11400
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.8887171561051005,
|
| 812 |
+
"grad_norm": 0.7135329842567444,
|
| 813 |
+
"learning_rate": 3.5195775373518807e-05,
|
| 814 |
+
"loss": 0.8883,
|
| 815 |
+
"step": 11500
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.8964451313755796,
|
| 819 |
+
"grad_norm": 0.9502096772193909,
|
| 820 |
+
"learning_rate": 3.5066975785677485e-05,
|
| 821 |
+
"loss": 0.8585,
|
| 822 |
+
"step": 11600
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9041731066460588,
|
| 826 |
+
"grad_norm": 1.4689842462539673,
|
| 827 |
+
"learning_rate": 3.493817619783617e-05,
|
| 828 |
+
"loss": 0.9306,
|
| 829 |
+
"step": 11700
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9119010819165378,
|
| 833 |
+
"grad_norm": 0.41735073924064636,
|
| 834 |
+
"learning_rate": 3.480937660999485e-05,
|
| 835 |
+
"loss": 0.8678,
|
| 836 |
+
"step": 11800
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.919629057187017,
|
| 840 |
+
"grad_norm": 0.6292606592178345,
|
| 841 |
+
"learning_rate": 3.4680577022153535e-05,
|
| 842 |
+
"loss": 0.8864,
|
| 843 |
+
"step": 11900
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.9273570324574961,
|
| 847 |
+
"grad_norm": 0.7318680882453918,
|
| 848 |
+
"learning_rate": 3.455177743431221e-05,
|
| 849 |
+
"loss": 0.8658,
|
| 850 |
+
"step": 12000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.9350850077279753,
|
| 854 |
+
"grad_norm": 1.8119789361953735,
|
| 855 |
+
"learning_rate": 3.442297784647089e-05,
|
| 856 |
+
"loss": 0.8236,
|
| 857 |
+
"step": 12100
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.9428129829984544,
|
| 861 |
+
"grad_norm": 2.469245195388794,
|
| 862 |
+
"learning_rate": 3.429417825862957e-05,
|
| 863 |
+
"loss": 0.9092,
|
| 864 |
+
"step": 12200
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.9505409582689336,
|
| 868 |
+
"grad_norm": 0.9728156924247742,
|
| 869 |
+
"learning_rate": 3.416537867078826e-05,
|
| 870 |
+
"loss": 0.9168,
|
| 871 |
+
"step": 12300
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.9582689335394127,
|
| 875 |
+
"grad_norm": 1.0292718410491943,
|
| 876 |
+
"learning_rate": 3.4036579082946936e-05,
|
| 877 |
+
"loss": 0.8535,
|
| 878 |
+
"step": 12400
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.9659969088098919,
|
| 882 |
+
"grad_norm": 1.157630443572998,
|
| 883 |
+
"learning_rate": 3.390777949510562e-05,
|
| 884 |
+
"loss": 0.8482,
|
| 885 |
+
"step": 12500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.973724884080371,
|
| 889 |
+
"grad_norm": 0.944819986820221,
|
| 890 |
+
"learning_rate": 3.3778979907264293e-05,
|
| 891 |
+
"loss": 0.9321,
|
| 892 |
+
"step": 12600
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.98145285935085,
|
| 896 |
+
"grad_norm": 0.8043965697288513,
|
| 897 |
+
"learning_rate": 3.365018031942298e-05,
|
| 898 |
+
"loss": 0.905,
|
| 899 |
+
"step": 12700
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.9891808346213292,
|
| 903 |
+
"grad_norm": 0.81541508436203,
|
| 904 |
+
"learning_rate": 3.352138073158166e-05,
|
| 905 |
+
"loss": 0.8518,
|
| 906 |
+
"step": 12800
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.9969088098918083,
|
| 910 |
+
"grad_norm": 0.9929108619689941,
|
| 911 |
+
"learning_rate": 3.339258114374034e-05,
|
| 912 |
+
"loss": 0.8439,
|
| 913 |
+
"step": 12900
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0046367851622875,
|
| 917 |
+
"grad_norm": 1.1995909214019775,
|
| 918 |
+
"learning_rate": 3.326378155589902e-05,
|
| 919 |
+
"loss": 0.8566,
|
| 920 |
+
"step": 13000
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.0123647604327666,
|
| 924 |
+
"grad_norm": 0.7053945064544678,
|
| 925 |
+
"learning_rate": 3.313498196805771e-05,
|
| 926 |
+
"loss": 0.8943,
|
| 927 |
+
"step": 13100
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.0200927357032457,
|
| 931 |
+
"grad_norm": 0.7920809984207153,
|
| 932 |
+
"learning_rate": 3.300618238021638e-05,
|
| 933 |
+
"loss": 0.8741,
|
| 934 |
+
"step": 13200
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.027820710973725,
|
| 938 |
+
"grad_norm": 1.0155510902404785,
|
| 939 |
+
"learning_rate": 3.2877382792375065e-05,
|
| 940 |
+
"loss": 0.8022,
|
| 941 |
+
"step": 13300
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.035548686244204,
|
| 945 |
+
"grad_norm": 1.0440579652786255,
|
| 946 |
+
"learning_rate": 3.2748583204533744e-05,
|
| 947 |
+
"loss": 0.8103,
|
| 948 |
+
"step": 13400
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.0432766615146831,
|
| 952 |
+
"grad_norm": 0.9690439105033875,
|
| 953 |
+
"learning_rate": 3.261978361669243e-05,
|
| 954 |
+
"loss": 0.8003,
|
| 955 |
+
"step": 13500
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.0510046367851622,
|
| 959 |
+
"grad_norm": 0.9963751435279846,
|
| 960 |
+
"learning_rate": 3.249098402885111e-05,
|
| 961 |
+
"loss": 0.8653,
|
| 962 |
+
"step": 13600
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.0587326120556415,
|
| 966 |
+
"grad_norm": 0.49158361554145813,
|
| 967 |
+
"learning_rate": 3.2362184441009794e-05,
|
| 968 |
+
"loss": 0.8642,
|
| 969 |
+
"step": 13700
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.0664605873261206,
|
| 973 |
+
"grad_norm": 0.9656476378440857,
|
| 974 |
+
"learning_rate": 3.2233384853168466e-05,
|
| 975 |
+
"loss": 0.8053,
|
| 976 |
+
"step": 13800
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.0741885625965997,
|
| 980 |
+
"grad_norm": 1.2530566453933716,
|
| 981 |
+
"learning_rate": 3.210458526532715e-05,
|
| 982 |
+
"loss": 0.8381,
|
| 983 |
+
"step": 13900
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.0819165378670788,
|
| 987 |
+
"grad_norm": 0.8346812725067139,
|
| 988 |
+
"learning_rate": 3.197578567748583e-05,
|
| 989 |
+
"loss": 0.8065,
|
| 990 |
+
"step": 14000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.089644513137558,
|
| 994 |
+
"grad_norm": 1.1836985349655151,
|
| 995 |
+
"learning_rate": 3.1846986089644516e-05,
|
| 996 |
+
"loss": 0.8483,
|
| 997 |
+
"step": 14100
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.0973724884080371,
|
| 1001 |
+
"grad_norm": 0.9917717576026917,
|
| 1002 |
+
"learning_rate": 3.1718186501803195e-05,
|
| 1003 |
+
"loss": 0.8459,
|
| 1004 |
+
"step": 14200
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.1051004636785162,
|
| 1008 |
+
"grad_norm": 1.0433543920516968,
|
| 1009 |
+
"learning_rate": 3.158938691396188e-05,
|
| 1010 |
+
"loss": 0.8446,
|
| 1011 |
+
"step": 14300
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.1128284389489953,
|
| 1015 |
+
"grad_norm": 1.1880654096603394,
|
| 1016 |
+
"learning_rate": 3.146058732612055e-05,
|
| 1017 |
+
"loss": 0.8122,
|
| 1018 |
+
"step": 14400
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.1205564142194744,
|
| 1022 |
+
"grad_norm": 0.46447160840034485,
|
| 1023 |
+
"learning_rate": 3.133178773827924e-05,
|
| 1024 |
+
"loss": 0.7701,
|
| 1025 |
+
"step": 14500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.1282843894899537,
|
| 1029 |
+
"grad_norm": 1.4256715774536133,
|
| 1030 |
+
"learning_rate": 3.1202988150437917e-05,
|
| 1031 |
+
"loss": 0.8635,
|
| 1032 |
+
"step": 14600
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.1360123647604328,
|
| 1036 |
+
"grad_norm": 0.9805368781089783,
|
| 1037 |
+
"learning_rate": 3.10741885625966e-05,
|
| 1038 |
+
"loss": 0.8218,
|
| 1039 |
+
"step": 14700
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.1437403400309119,
|
| 1043 |
+
"grad_norm": 1.1528363227844238,
|
| 1044 |
+
"learning_rate": 3.094538897475529e-05,
|
| 1045 |
+
"loss": 0.8366,
|
| 1046 |
+
"step": 14800
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.1514683153013912,
|
| 1050 |
+
"grad_norm": 0.9413203001022339,
|
| 1051 |
+
"learning_rate": 3.0816589386913966e-05,
|
| 1052 |
+
"loss": 0.8323,
|
| 1053 |
+
"step": 14900
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.1591962905718702,
|
| 1057 |
+
"grad_norm": 0.9010465145111084,
|
| 1058 |
+
"learning_rate": 3.0687789799072645e-05,
|
| 1059 |
+
"loss": 0.8863,
|
| 1060 |
+
"step": 15000
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.1669242658423493,
|
| 1064 |
+
"grad_norm": 0.8075920343399048,
|
| 1065 |
+
"learning_rate": 3.0558990211231324e-05,
|
| 1066 |
+
"loss": 0.8015,
|
| 1067 |
+
"step": 15100
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.1746522411128284,
|
| 1071 |
+
"grad_norm": 0.7727937698364258,
|
| 1072 |
+
"learning_rate": 3.043019062339001e-05,
|
| 1073 |
+
"loss": 0.822,
|
| 1074 |
+
"step": 15200
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.1823802163833075,
|
| 1078 |
+
"grad_norm": 0.8842815160751343,
|
| 1079 |
+
"learning_rate": 3.030139103554869e-05,
|
| 1080 |
+
"loss": 0.8069,
|
| 1081 |
+
"step": 15300
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.1901081916537868,
|
| 1085 |
+
"grad_norm": 1.687024474143982,
|
| 1086 |
+
"learning_rate": 3.017259144770737e-05,
|
| 1087 |
+
"loss": 0.8723,
|
| 1088 |
+
"step": 15400
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.1978361669242659,
|
| 1092 |
+
"grad_norm": 0.9511739015579224,
|
| 1093 |
+
"learning_rate": 3.004379185986605e-05,
|
| 1094 |
+
"loss": 0.7482,
|
| 1095 |
+
"step": 15500
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.205564142194745,
|
| 1099 |
+
"grad_norm": 1.404523253440857,
|
| 1100 |
+
"learning_rate": 2.9914992272024735e-05,
|
| 1101 |
+
"loss": 0.8435,
|
| 1102 |
+
"step": 15600
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.213292117465224,
|
| 1106 |
+
"grad_norm": 1.1667975187301636,
|
| 1107 |
+
"learning_rate": 2.978619268418341e-05,
|
| 1108 |
+
"loss": 0.8717,
|
| 1109 |
+
"step": 15700
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.2210200927357033,
|
| 1113 |
+
"grad_norm": 1.286392331123352,
|
| 1114 |
+
"learning_rate": 2.9657393096342096e-05,
|
| 1115 |
+
"loss": 0.8148,
|
| 1116 |
+
"step": 15800
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.2287480680061824,
|
| 1120 |
+
"grad_norm": 0.915986955165863,
|
| 1121 |
+
"learning_rate": 2.9528593508500775e-05,
|
| 1122 |
+
"loss": 0.8285,
|
| 1123 |
+
"step": 15900
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.2364760432766615,
|
| 1127 |
+
"grad_norm": 0.9131502509117126,
|
| 1128 |
+
"learning_rate": 2.9399793920659457e-05,
|
| 1129 |
+
"loss": 0.8686,
|
| 1130 |
+
"step": 16000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.2442040185471406,
|
| 1134 |
+
"grad_norm": 1.6465792655944824,
|
| 1135 |
+
"learning_rate": 2.9270994332818136e-05,
|
| 1136 |
+
"loss": 0.7989,
|
| 1137 |
+
"step": 16100
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.2519319938176197,
|
| 1141 |
+
"grad_norm": 0.6225204467773438,
|
| 1142 |
+
"learning_rate": 2.914219474497682e-05,
|
| 1143 |
+
"loss": 0.8871,
|
| 1144 |
+
"step": 16200
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.259659969088099,
|
| 1148 |
+
"grad_norm": 1.4327455759048462,
|
| 1149 |
+
"learning_rate": 2.9013395157135496e-05,
|
| 1150 |
+
"loss": 0.8531,
|
| 1151 |
+
"step": 16300
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.267387944358578,
|
| 1155 |
+
"grad_norm": 1.1303315162658691,
|
| 1156 |
+
"learning_rate": 2.8884595569294182e-05,
|
| 1157 |
+
"loss": 0.8464,
|
| 1158 |
+
"step": 16400
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.2751159196290571,
|
| 1162 |
+
"grad_norm": 0.8437920212745667,
|
| 1163 |
+
"learning_rate": 2.875579598145286e-05,
|
| 1164 |
+
"loss": 0.8394,
|
| 1165 |
+
"step": 16500
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.2828438948995364,
|
| 1169 |
+
"grad_norm": 1.1533674001693726,
|
| 1170 |
+
"learning_rate": 2.8626996393611543e-05,
|
| 1171 |
+
"loss": 0.8548,
|
| 1172 |
+
"step": 16600
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.2905718701700155,
|
| 1176 |
+
"grad_norm": 0.858174204826355,
|
| 1177 |
+
"learning_rate": 2.8498196805770222e-05,
|
| 1178 |
+
"loss": 0.8281,
|
| 1179 |
+
"step": 16700
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.2982998454404946,
|
| 1183 |
+
"grad_norm": 0.8084927797317505,
|
| 1184 |
+
"learning_rate": 2.8369397217928907e-05,
|
| 1185 |
+
"loss": 0.8215,
|
| 1186 |
+
"step": 16800
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.3060278207109737,
|
| 1190 |
+
"grad_norm": 1.1206783056259155,
|
| 1191 |
+
"learning_rate": 2.8240597630087583e-05,
|
| 1192 |
+
"loss": 0.8648,
|
| 1193 |
+
"step": 16900
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.3137557959814528,
|
| 1197 |
+
"grad_norm": 1.336126685142517,
|
| 1198 |
+
"learning_rate": 2.8111798042246268e-05,
|
| 1199 |
+
"loss": 0.8386,
|
| 1200 |
+
"step": 17000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.321483771251932,
|
| 1204 |
+
"grad_norm": 1.350467324256897,
|
| 1205 |
+
"learning_rate": 2.798428645028336e-05,
|
| 1206 |
+
"loss": 0.7863,
|
| 1207 |
+
"step": 17100
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.3292117465224111,
|
| 1211 |
+
"grad_norm": 1.1663061380386353,
|
| 1212 |
+
"learning_rate": 2.785548686244204e-05,
|
| 1213 |
+
"loss": 0.8944,
|
| 1214 |
+
"step": 17200
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.3369397217928902,
|
| 1218 |
+
"grad_norm": 1.4521268606185913,
|
| 1219 |
+
"learning_rate": 2.7727975270479132e-05,
|
| 1220 |
+
"loss": 0.8633,
|
| 1221 |
+
"step": 17300
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.3446676970633695,
|
| 1225 |
+
"grad_norm": 1.0562511682510376,
|
| 1226 |
+
"learning_rate": 2.7599175682637818e-05,
|
| 1227 |
+
"loss": 0.8684,
|
| 1228 |
+
"step": 17400
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.3523956723338486,
|
| 1232 |
+
"grad_norm": 0.7794145941734314,
|
| 1233 |
+
"learning_rate": 2.7470376094796497e-05,
|
| 1234 |
+
"loss": 0.8334,
|
| 1235 |
+
"step": 17500
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.3601236476043277,
|
| 1239 |
+
"grad_norm": 1.1900535821914673,
|
| 1240 |
+
"learning_rate": 2.734157650695518e-05,
|
| 1241 |
+
"loss": 0.8278,
|
| 1242 |
+
"step": 17600
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.3678516228748068,
|
| 1246 |
+
"grad_norm": 1.2511202096939087,
|
| 1247 |
+
"learning_rate": 2.7212776919113858e-05,
|
| 1248 |
+
"loss": 0.8056,
|
| 1249 |
+
"step": 17700
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.3755795981452859,
|
| 1253 |
+
"grad_norm": 0.5958196520805359,
|
| 1254 |
+
"learning_rate": 2.7083977331272543e-05,
|
| 1255 |
+
"loss": 0.7903,
|
| 1256 |
+
"step": 17800
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.383307573415765,
|
| 1260 |
+
"grad_norm": 0.9310637712478638,
|
| 1261 |
+
"learning_rate": 2.695517774343122e-05,
|
| 1262 |
+
"loss": 0.806,
|
| 1263 |
+
"step": 17900
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.3910355486862442,
|
| 1267 |
+
"grad_norm": 0.8104686737060547,
|
| 1268 |
+
"learning_rate": 2.6826378155589904e-05,
|
| 1269 |
+
"loss": 0.8067,
|
| 1270 |
+
"step": 18000
|
| 1271 |
+
}
|
| 1272 |
+
],
|
| 1273 |
+
"logging_steps": 100,
|
| 1274 |
+
"max_steps": 38820,
|
| 1275 |
+
"num_input_tokens_seen": 0,
|
| 1276 |
+
"num_train_epochs": 3,
|
| 1277 |
+
"save_steps": 500,
|
| 1278 |
+
"stateful_callbacks": {
|
| 1279 |
+
"TrainerControl": {
|
| 1280 |
+
"args": {
|
| 1281 |
+
"should_epoch_stop": false,
|
| 1282 |
+
"should_evaluate": false,
|
| 1283 |
+
"should_log": false,
|
| 1284 |
+
"should_save": true,
|
| 1285 |
+
"should_training_stop": false
|
| 1286 |
+
},
|
| 1287 |
+
"attributes": {}
|
| 1288 |
+
}
|
| 1289 |
+
},
|
| 1290 |
+
"total_flos": 1.8813026304e+16,
|
| 1291 |
+
"train_batch_size": 4,
|
| 1292 |
+
"trial_name": null,
|
| 1293 |
+
"trial_params": null
|
| 1294 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:335c2996c101370d2f40ae68b10dc4705207dda5805f34ebb386ffc4e05d1b83
|
| 3 |
+
size 5304
|