Upload folder using huggingface_hub
Browse files- config.json +39 -0
- generation_config.json +6 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2157 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 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 |
+
"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 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.48.1",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50257
|
| 39 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.48.1"
|
| 6 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e87f069c0839be87b776f61e5d0cf2e63b39ea44a4c887c24c80cb45bca095a5
|
| 3 |
+
size 995642298
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82f31ef929a6cde0411b883525b7d28544563ee3283fba2f86445dbec0bbd8e9
|
| 3 |
+
size 497807706
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:476f620691243aaa7324010380016f742705e0661686a455731fd5bbc35b158a
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8d0ae2d00df278ca1c951f36de6151714875cca07d377b45461ba9698f751e6
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2157 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.016333938294010888,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 5.4446460980036295e-05,
|
| 13 |
+
"grad_norm": 2.606283664703369,
|
| 14 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 15 |
+
"loss": 3.3597,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00010889292196007259,
|
| 20 |
+
"grad_norm": 2.6963095664978027,
|
| 21 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 22 |
+
"loss": 3.3809,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0001633393829401089,
|
| 27 |
+
"grad_norm": 3.531561851501465,
|
| 28 |
+
"learning_rate": 8.999999999999999e-05,
|
| 29 |
+
"loss": 3.353,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.00021778584392014518,
|
| 34 |
+
"grad_norm": 2.844101667404175,
|
| 35 |
+
"learning_rate": 0.00011999999999999999,
|
| 36 |
+
"loss": 3.3358,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0002722323049001815,
|
| 41 |
+
"grad_norm": 2.73467755317688,
|
| 42 |
+
"learning_rate": 0.00015,
|
| 43 |
+
"loss": 3.2899,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0003266787658802178,
|
| 48 |
+
"grad_norm": 2.266099452972412,
|
| 49 |
+
"learning_rate": 0.00017999999999999998,
|
| 50 |
+
"loss": 3.2718,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.00038112522686025406,
|
| 55 |
+
"grad_norm": 1.782454490661621,
|
| 56 |
+
"learning_rate": 0.00020999999999999998,
|
| 57 |
+
"loss": 3.25,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.00043557168784029036,
|
| 62 |
+
"grad_norm": 1.6878714561462402,
|
| 63 |
+
"learning_rate": 0.00023999999999999998,
|
| 64 |
+
"loss": 3.2508,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0004900181488203267,
|
| 69 |
+
"grad_norm": 1.656855821609497,
|
| 70 |
+
"learning_rate": 0.00027,
|
| 71 |
+
"loss": 3.2482,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.000544464609800363,
|
| 76 |
+
"grad_norm": 1.196555256843567,
|
| 77 |
+
"learning_rate": 0.0003,
|
| 78 |
+
"loss": 3.2249,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0005989110707803993,
|
| 83 |
+
"grad_norm": 0.7616098523139954,
|
| 84 |
+
"learning_rate": 0.00033,
|
| 85 |
+
"loss": 3.2238,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0006533575317604356,
|
| 90 |
+
"grad_norm": 0.8151504993438721,
|
| 91 |
+
"learning_rate": 0.00035999999999999997,
|
| 92 |
+
"loss": 3.1822,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0007078039927404719,
|
| 97 |
+
"grad_norm": 0.7040852308273315,
|
| 98 |
+
"learning_rate": 0.00039,
|
| 99 |
+
"loss": 3.1587,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0007622504537205081,
|
| 104 |
+
"grad_norm": 0.5113953351974487,
|
| 105 |
+
"learning_rate": 0.00041999999999999996,
|
| 106 |
+
"loss": 3.2203,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.0008166969147005445,
|
| 111 |
+
"grad_norm": 0.6095477342605591,
|
| 112 |
+
"learning_rate": 0.00045,
|
| 113 |
+
"loss": 3.1912,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0008711433756805807,
|
| 118 |
+
"grad_norm": 0.7497212290763855,
|
| 119 |
+
"learning_rate": 0.00047999999999999996,
|
| 120 |
+
"loss": 3.1985,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0009255898366606171,
|
| 125 |
+
"grad_norm": 0.9816092848777771,
|
| 126 |
+
"learning_rate": 0.0005099999999999999,
|
| 127 |
+
"loss": 3.1859,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0009800362976406533,
|
| 132 |
+
"grad_norm": 0.9469059109687805,
|
| 133 |
+
"learning_rate": 0.00054,
|
| 134 |
+
"loss": 3.2167,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0010344827586206897,
|
| 139 |
+
"grad_norm": 0.6933241486549377,
|
| 140 |
+
"learning_rate": 0.00057,
|
| 141 |
+
"loss": 3.1399,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.001088929219600726,
|
| 146 |
+
"grad_norm": 0.9282199144363403,
|
| 147 |
+
"learning_rate": 0.0006,
|
| 148 |
+
"loss": 3.1684,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0011433756805807622,
|
| 153 |
+
"grad_norm": 0.48486050963401794,
|
| 154 |
+
"learning_rate": 0.000599990558513205,
|
| 155 |
+
"loss": 3.2108,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0011978221415607985,
|
| 160 |
+
"grad_norm": 0.5165353417396545,
|
| 161 |
+
"learning_rate": 0.000599962235241376,
|
| 162 |
+
"loss": 3.2203,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.001252268602540835,
|
| 167 |
+
"grad_norm": 0.40520766377449036,
|
| 168 |
+
"learning_rate": 0.0005999150337500299,
|
| 169 |
+
"loss": 3.1709,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0013067150635208712,
|
| 174 |
+
"grad_norm": 0.4057173430919647,
|
| 175 |
+
"learning_rate": 0.0005998489599811971,
|
| 176 |
+
"loss": 3.1699,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0013611615245009074,
|
| 181 |
+
"grad_norm": 0.3643300533294678,
|
| 182 |
+
"learning_rate": 0.0005997640222526724,
|
| 183 |
+
"loss": 3.1722,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0014156079854809437,
|
| 188 |
+
"grad_norm": 0.4470251798629761,
|
| 189 |
+
"learning_rate": 0.0005996602312569683,
|
| 190 |
+
"loss": 3.1321,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.00147005444646098,
|
| 195 |
+
"grad_norm": 0.3419328033924103,
|
| 196 |
+
"learning_rate": 0.0005995376000599692,
|
| 197 |
+
"loss": 3.1753,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.0015245009074410162,
|
| 202 |
+
"grad_norm": 0.4055408537387848,
|
| 203 |
+
"learning_rate": 0.0005993961440992859,
|
| 204 |
+
"loss": 3.1604,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.0015789473684210526,
|
| 209 |
+
"grad_norm": 0.3212841749191284,
|
| 210 |
+
"learning_rate": 0.0005992358811823128,
|
| 211 |
+
"loss": 3.1666,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.001633393829401089,
|
| 216 |
+
"grad_norm": 0.30549293756484985,
|
| 217 |
+
"learning_rate": 0.0005990568314839863,
|
| 218 |
+
"loss": 3.1292,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.0016878402903811253,
|
| 223 |
+
"grad_norm": 0.3610576391220093,
|
| 224 |
+
"learning_rate": 0.0005988590175442446,
|
| 225 |
+
"loss": 3.1435,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.0017422867513611614,
|
| 230 |
+
"grad_norm": 0.2855391502380371,
|
| 231 |
+
"learning_rate": 0.0005986424642651901,
|
| 232 |
+
"loss": 3.1432,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.0017967332123411978,
|
| 237 |
+
"grad_norm": 0.30511122941970825,
|
| 238 |
+
"learning_rate": 0.0005984071989079554,
|
| 239 |
+
"loss": 3.1463,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.0018511796733212342,
|
| 244 |
+
"grad_norm": 0.29414448142051697,
|
| 245 |
+
"learning_rate": 0.0005981532510892706,
|
| 246 |
+
"loss": 3.1336,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.0019056261343012705,
|
| 251 |
+
"grad_norm": 0.3027634620666504,
|
| 252 |
+
"learning_rate": 0.0005978806527777354,
|
| 253 |
+
"loss": 3.1277,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.0019600725952813067,
|
| 258 |
+
"grad_norm": 0.3479376435279846,
|
| 259 |
+
"learning_rate": 0.0005975894382897944,
|
| 260 |
+
"loss": 3.1298,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.002014519056261343,
|
| 265 |
+
"grad_norm": 0.31864309310913086,
|
| 266 |
+
"learning_rate": 0.0005972796442854177,
|
| 267 |
+
"loss": 3.1501,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.0020689655172413794,
|
| 272 |
+
"grad_norm": 0.33572497963905334,
|
| 273 |
+
"learning_rate": 0.0005969513097634852,
|
| 274 |
+
"loss": 3.1505,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.0021234119782214157,
|
| 279 |
+
"grad_norm": 0.23914705216884613,
|
| 280 |
+
"learning_rate": 0.0005966044760568779,
|
| 281 |
+
"loss": 3.1518,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.002177858439201452,
|
| 286 |
+
"grad_norm": 0.34076428413391113,
|
| 287 |
+
"learning_rate": 0.0005962391868272735,
|
| 288 |
+
"loss": 3.1378,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.002232304900181488,
|
| 293 |
+
"grad_norm": 0.25502562522888184,
|
| 294 |
+
"learning_rate": 0.0005958554880596515,
|
| 295 |
+
"loss": 3.1532,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.0022867513611615244,
|
| 300 |
+
"grad_norm": 0.2586739659309387,
|
| 301 |
+
"learning_rate": 0.000595453428056503,
|
| 302 |
+
"loss": 3.1261,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.0023411978221415607,
|
| 307 |
+
"grad_norm": 0.3028804659843445,
|
| 308 |
+
"learning_rate": 0.0005950330574317509,
|
| 309 |
+
"loss": 3.1535,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.002395644283121597,
|
| 314 |
+
"grad_norm": 0.2421780228614807,
|
| 315 |
+
"learning_rate": 0.0005945944291043779,
|
| 316 |
+
"loss": 3.1712,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.0024500907441016334,
|
| 321 |
+
"grad_norm": 0.2662357687950134,
|
| 322 |
+
"learning_rate": 0.0005941375982917649,
|
| 323 |
+
"loss": 3.1486,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.00250453720508167,
|
| 328 |
+
"grad_norm": 0.23532505333423615,
|
| 329 |
+
"learning_rate": 0.0005936626225027395,
|
| 330 |
+
"loss": 3.1478,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.002558983666061706,
|
| 335 |
+
"grad_norm": 0.33855125308036804,
|
| 336 |
+
"learning_rate": 0.000593169561530337,
|
| 337 |
+
"loss": 3.1025,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.0026134301270417425,
|
| 342 |
+
"grad_norm": 0.23862820863723755,
|
| 343 |
+
"learning_rate": 0.000592658477444273,
|
| 344 |
+
"loss": 3.1272,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.0026678765880217784,
|
| 349 |
+
"grad_norm": 0.2811816930770874,
|
| 350 |
+
"learning_rate": 0.0005921294345831293,
|
| 351 |
+
"loss": 3.1428,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.0027223230490018148,
|
| 356 |
+
"grad_norm": 0.2487388700246811,
|
| 357 |
+
"learning_rate": 0.0005915824995462552,
|
| 358 |
+
"loss": 3.1305,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.002776769509981851,
|
| 363 |
+
"grad_norm": 0.2535880506038666,
|
| 364 |
+
"learning_rate": 0.0005910177411853828,
|
| 365 |
+
"loss": 3.1648,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.0028312159709618875,
|
| 370 |
+
"grad_norm": 0.3038806617259979,
|
| 371 |
+
"learning_rate": 0.0005904352305959605,
|
| 372 |
+
"loss": 3.1285,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.002885662431941924,
|
| 377 |
+
"grad_norm": 0.23585520684719086,
|
| 378 |
+
"learning_rate": 0.000589835041108202,
|
| 379 |
+
"loss": 3.1428,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.00294010889292196,
|
| 384 |
+
"grad_norm": 0.2504676580429077,
|
| 385 |
+
"learning_rate": 0.0005892172482778558,
|
| 386 |
+
"loss": 3.0991,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.0029945553539019966,
|
| 391 |
+
"grad_norm": 0.2501893937587738,
|
| 392 |
+
"learning_rate": 0.000588581929876693,
|
| 393 |
+
"loss": 3.1339,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.0030490018148820325,
|
| 398 |
+
"grad_norm": 0.2685960531234741,
|
| 399 |
+
"learning_rate": 0.0005879291658827176,
|
| 400 |
+
"loss": 3.1412,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.003103448275862069,
|
| 405 |
+
"grad_norm": 0.23503637313842773,
|
| 406 |
+
"learning_rate": 0.0005872590384700979,
|
| 407 |
+
"loss": 3.1625,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.003157894736842105,
|
| 412 |
+
"grad_norm": 0.24152350425720215,
|
| 413 |
+
"learning_rate": 0.0005865716319988223,
|
| 414 |
+
"loss": 3.1059,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.0032123411978221415,
|
| 419 |
+
"grad_norm": 0.20929178595542908,
|
| 420 |
+
"learning_rate": 0.000585867033004079,
|
| 421 |
+
"loss": 3.1307,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.003266787658802178,
|
| 426 |
+
"grad_norm": 0.2642001807689667,
|
| 427 |
+
"learning_rate": 0.0005851453301853628,
|
| 428 |
+
"loss": 3.1134,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.0033212341197822143,
|
| 433 |
+
"grad_norm": 0.24187405407428741,
|
| 434 |
+
"learning_rate": 0.0005844066143953087,
|
| 435 |
+
"loss": 3.1523,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.0033756805807622506,
|
| 440 |
+
"grad_norm": 0.2009871006011963,
|
| 441 |
+
"learning_rate": 0.0005836509786282552,
|
| 442 |
+
"loss": 3.1294,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.0034301270417422865,
|
| 447 |
+
"grad_norm": 0.25516337156295776,
|
| 448 |
+
"learning_rate": 0.000582878518008537,
|
| 449 |
+
"loss": 3.1309,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.003484573502722323,
|
| 454 |
+
"grad_norm": 0.22489990293979645,
|
| 455 |
+
"learning_rate": 0.0005820893297785106,
|
| 456 |
+
"loss": 3.1414,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.0035390199637023593,
|
| 461 |
+
"grad_norm": 0.22305169701576233,
|
| 462 |
+
"learning_rate": 0.000581283513286313,
|
| 463 |
+
"loss": 3.1573,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.0035934664246823956,
|
| 468 |
+
"grad_norm": 0.21161314845085144,
|
| 469 |
+
"learning_rate": 0.0005804611699733543,
|
| 470 |
+
"loss": 3.1103,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.003647912885662432,
|
| 475 |
+
"grad_norm": 0.2384558469057083,
|
| 476 |
+
"learning_rate": 0.0005796224033615482,
|
| 477 |
+
"loss": 3.1364,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.0037023593466424683,
|
| 482 |
+
"grad_norm": 0.2701220214366913,
|
| 483 |
+
"learning_rate": 0.0005787673190402799,
|
| 484 |
+
"loss": 3.1369,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.0037568058076225047,
|
| 489 |
+
"grad_norm": 0.19185633957386017,
|
| 490 |
+
"learning_rate": 0.0005778960246531138,
|
| 491 |
+
"loss": 3.1247,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.003811252268602541,
|
| 496 |
+
"grad_norm": 0.3245127499103546,
|
| 497 |
+
"learning_rate": 0.0005770086298842426,
|
| 498 |
+
"loss": 3.126,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.003865698729582577,
|
| 503 |
+
"grad_norm": 0.281764417886734,
|
| 504 |
+
"learning_rate": 0.0005761052464446795,
|
| 505 |
+
"loss": 3.1295,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.003920145190562613,
|
| 510 |
+
"grad_norm": 0.24820677936077118,
|
| 511 |
+
"learning_rate": 0.0005751859880581954,
|
| 512 |
+
"loss": 3.1337,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.00397459165154265,
|
| 517 |
+
"grad_norm": 0.24963483214378357,
|
| 518 |
+
"learning_rate": 0.0005742509704470024,
|
| 519 |
+
"loss": 3.119,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.004029038112522686,
|
| 524 |
+
"grad_norm": 0.264715313911438,
|
| 525 |
+
"learning_rate": 0.0005733003113171864,
|
| 526 |
+
"loss": 3.1203,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.004083484573502722,
|
| 531 |
+
"grad_norm": 0.27801015973091125,
|
| 532 |
+
"learning_rate": 0.000572334130343889,
|
| 533 |
+
"loss": 3.1531,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.004137931034482759,
|
| 538 |
+
"grad_norm": 0.21041367948055267,
|
| 539 |
+
"learning_rate": 0.0005713525491562421,
|
| 540 |
+
"loss": 3.1208,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.004192377495462795,
|
| 545 |
+
"grad_norm": 0.22113123536109924,
|
| 546 |
+
"learning_rate": 0.0005703556913220566,
|
| 547 |
+
"loss": 3.1209,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.0042468239564428314,
|
| 552 |
+
"grad_norm": 0.22394251823425293,
|
| 553 |
+
"learning_rate": 0.0005693436823322671,
|
| 554 |
+
"loss": 3.1101,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.004301270417422867,
|
| 559 |
+
"grad_norm": 0.26046207547187805,
|
| 560 |
+
"learning_rate": 0.0005683166495851336,
|
| 561 |
+
"loss": 3.0935,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.004355716878402904,
|
| 566 |
+
"grad_norm": 0.19576989114284515,
|
| 567 |
+
"learning_rate": 0.0005672747223702044,
|
| 568 |
+
"loss": 3.1381,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.00441016333938294,
|
| 573 |
+
"grad_norm": 0.22762754559516907,
|
| 574 |
+
"learning_rate": 0.0005662180318520402,
|
| 575 |
+
"loss": 3.1117,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.004464609800362976,
|
| 580 |
+
"grad_norm": 0.218556210398674,
|
| 581 |
+
"learning_rate": 0.0005651467110537016,
|
| 582 |
+
"loss": 3.1164,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.004519056261343013,
|
| 587 |
+
"grad_norm": 0.23464246094226837,
|
| 588 |
+
"learning_rate": 0.0005640608948400046,
|
| 589 |
+
"loss": 3.1533,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.004573502722323049,
|
| 594 |
+
"grad_norm": 0.23230504989624023,
|
| 595 |
+
"learning_rate": 0.0005629607199005416,
|
| 596 |
+
"loss": 3.128,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.0046279491833030855,
|
| 601 |
+
"grad_norm": 0.2193627804517746,
|
| 602 |
+
"learning_rate": 0.0005618463247324748,
|
| 603 |
+
"loss": 3.1162,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.004682395644283121,
|
| 608 |
+
"grad_norm": 0.22908492386341095,
|
| 609 |
+
"learning_rate": 0.0005607178496231011,
|
| 610 |
+
"loss": 3.1198,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.004736842105263158,
|
| 615 |
+
"grad_norm": 0.2191692590713501,
|
| 616 |
+
"learning_rate": 0.0005595754366321915,
|
| 617 |
+
"loss": 3.1135,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.004791288566243194,
|
| 622 |
+
"grad_norm": 0.21760721504688263,
|
| 623 |
+
"learning_rate": 0.0005584192295741086,
|
| 624 |
+
"loss": 3.112,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.00484573502722323,
|
| 629 |
+
"grad_norm": 0.27237263321876526,
|
| 630 |
+
"learning_rate": 0.0005572493739997012,
|
| 631 |
+
"loss": 3.1167,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.004900181488203267,
|
| 636 |
+
"grad_norm": 0.2498655915260315,
|
| 637 |
+
"learning_rate": 0.000556066017177982,
|
| 638 |
+
"loss": 3.119,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.004954627949183303,
|
| 643 |
+
"grad_norm": 0.23274663090705872,
|
| 644 |
+
"learning_rate": 0.0005548693080775885,
|
| 645 |
+
"loss": 3.168,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.00500907441016334,
|
| 650 |
+
"grad_norm": 0.22452446818351746,
|
| 651 |
+
"learning_rate": 0.0005536593973480297,
|
| 652 |
+
"loss": 3.1238,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.0050635208711433755,
|
| 657 |
+
"grad_norm": 0.2021019011735916,
|
| 658 |
+
"learning_rate": 0.000552436437300721,
|
| 659 |
+
"loss": 3.1363,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.005117967332123412,
|
| 664 |
+
"grad_norm": 0.20027242600917816,
|
| 665 |
+
"learning_rate": 0.0005512005818898111,
|
| 666 |
+
"loss": 3.1075,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.005172413793103448,
|
| 671 |
+
"grad_norm": 0.20116809010505676,
|
| 672 |
+
"learning_rate": 0.0005499519866928005,
|
| 673 |
+
"loss": 3.1137,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.005226860254083485,
|
| 678 |
+
"grad_norm": 0.29089683294296265,
|
| 679 |
+
"learning_rate": 0.0005486908088909568,
|
| 680 |
+
"loss": 3.1435,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.005281306715063521,
|
| 685 |
+
"grad_norm": 0.26022160053253174,
|
| 686 |
+
"learning_rate": 0.0005474172072495275,
|
| 687 |
+
"loss": 3.1348,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.005335753176043557,
|
| 692 |
+
"grad_norm": 0.25501441955566406,
|
| 693 |
+
"learning_rate": 0.0005461313420977536,
|
| 694 |
+
"loss": 3.1243,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.005390199637023594,
|
| 699 |
+
"grad_norm": 0.2464355230331421,
|
| 700 |
+
"learning_rate": 0.0005448333753086864,
|
| 701 |
+
"loss": 3.1329,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.0054446460980036296,
|
| 706 |
+
"grad_norm": 0.25668638944625854,
|
| 707 |
+
"learning_rate": 0.00054352347027881,
|
| 708 |
+
"loss": 3.0924,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.0054446460980036296,
|
| 713 |
+
"eval_loss": 3.0220067501068115,
|
| 714 |
+
"eval_runtime": 69.1667,
|
| 715 |
+
"eval_samples_per_second": 62.385,
|
| 716 |
+
"eval_steps_per_second": 15.6,
|
| 717 |
+
"step": 100
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.005499092558983666,
|
| 721 |
+
"grad_norm": 0.25321197509765625,
|
| 722 |
+
"learning_rate": 0.0005422017919074715,
|
| 723 |
+
"loss": 3.1289,
|
| 724 |
+
"step": 101
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.005553539019963702,
|
| 728 |
+
"grad_norm": 0.21445196866989136,
|
| 729 |
+
"learning_rate": 0.0005408685065761229,
|
| 730 |
+
"loss": 3.123,
|
| 731 |
+
"step": 102
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.005607985480943739,
|
| 735 |
+
"grad_norm": 0.227112278342247,
|
| 736 |
+
"learning_rate": 0.0005395237821273755,
|
| 737 |
+
"loss": 3.1302,
|
| 738 |
+
"step": 103
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.005662431941923775,
|
| 742 |
+
"grad_norm": 0.22493597865104675,
|
| 743 |
+
"learning_rate": 0.000538167787843871,
|
| 744 |
+
"loss": 3.1471,
|
| 745 |
+
"step": 104
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.005716878402903811,
|
| 749 |
+
"grad_norm": 0.24361659586429596,
|
| 750 |
+
"learning_rate": 0.0005368006944269708,
|
| 751 |
+
"loss": 3.1271,
|
| 752 |
+
"step": 105
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.005771324863883848,
|
| 756 |
+
"grad_norm": 0.2216707319021225,
|
| 757 |
+
"learning_rate": 0.0005354226739752678,
|
| 758 |
+
"loss": 3.1156,
|
| 759 |
+
"step": 106
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.005825771324863884,
|
| 763 |
+
"grad_norm": 0.2140239179134369,
|
| 764 |
+
"learning_rate": 0.0005340338999629203,
|
| 765 |
+
"loss": 3.1071,
|
| 766 |
+
"step": 107
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.00588021778584392,
|
| 770 |
+
"grad_norm": 0.22374412417411804,
|
| 771 |
+
"learning_rate": 0.0005326345472178154,
|
| 772 |
+
"loss": 3.1244,
|
| 773 |
+
"step": 108
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.005934664246823956,
|
| 777 |
+
"grad_norm": 0.21664555370807648,
|
| 778 |
+
"learning_rate": 0.0005312247918995588,
|
| 779 |
+
"loss": 3.1254,
|
| 780 |
+
"step": 109
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.005989110707803993,
|
| 784 |
+
"grad_norm": 0.21399718523025513,
|
| 785 |
+
"learning_rate": 0.0005298048114773004,
|
| 786 |
+
"loss": 3.119,
|
| 787 |
+
"step": 110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.006043557168784029,
|
| 791 |
+
"grad_norm": 0.18971006572246552,
|
| 792 |
+
"learning_rate": 0.0005283747847073922,
|
| 793 |
+
"loss": 3.1254,
|
| 794 |
+
"step": 111
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.006098003629764065,
|
| 798 |
+
"grad_norm": 0.1922215223312378,
|
| 799 |
+
"learning_rate": 0.0005269348916108859,
|
| 800 |
+
"loss": 3.1016,
|
| 801 |
+
"step": 112
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.006152450090744102,
|
| 805 |
+
"grad_norm": 0.21366891264915466,
|
| 806 |
+
"learning_rate": 0.00052548531345087,
|
| 807 |
+
"loss": 3.1192,
|
| 808 |
+
"step": 113
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.006206896551724138,
|
| 812 |
+
"grad_norm": 0.1934356689453125,
|
| 813 |
+
"learning_rate": 0.000524026232709652,
|
| 814 |
+
"loss": 3.1254,
|
| 815 |
+
"step": 114
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.0062613430127041745,
|
| 819 |
+
"grad_norm": 0.23197585344314575,
|
| 820 |
+
"learning_rate": 0.0005225578330657859,
|
| 821 |
+
"loss": 3.1503,
|
| 822 |
+
"step": 115
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.00631578947368421,
|
| 826 |
+
"grad_norm": 0.227361798286438,
|
| 827 |
+
"learning_rate": 0.0005210802993709497,
|
| 828 |
+
"loss": 3.112,
|
| 829 |
+
"step": 116
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.006370235934664247,
|
| 833 |
+
"grad_norm": 0.25826945900917053,
|
| 834 |
+
"learning_rate": 0.0005195938176266751,
|
| 835 |
+
"loss": 3.1405,
|
| 836 |
+
"step": 117
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.006424682395644283,
|
| 840 |
+
"grad_norm": 0.2799200117588043,
|
| 841 |
+
"learning_rate": 0.000518098574960932,
|
| 842 |
+
"loss": 3.1425,
|
| 843 |
+
"step": 118
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.006479128856624319,
|
| 847 |
+
"grad_norm": 0.24563385546207428,
|
| 848 |
+
"learning_rate": 0.0005165947596045723,
|
| 849 |
+
"loss": 3.1573,
|
| 850 |
+
"step": 119
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.006533575317604356,
|
| 854 |
+
"grad_norm": 0.2642490267753601,
|
| 855 |
+
"learning_rate": 0.0005150825608676336,
|
| 856 |
+
"loss": 3.1267,
|
| 857 |
+
"step": 120
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.006588021778584392,
|
| 861 |
+
"grad_norm": 0.2301069051027298,
|
| 862 |
+
"learning_rate": 0.0005135621691155083,
|
| 863 |
+
"loss": 3.1151,
|
| 864 |
+
"step": 121
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.0066424682395644285,
|
| 868 |
+
"grad_norm": 0.22567883133888245,
|
| 869 |
+
"learning_rate": 0.0005120337757449781,
|
| 870 |
+
"loss": 3.1465,
|
| 871 |
+
"step": 122
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.0066969147005444644,
|
| 875 |
+
"grad_norm": 0.23506537079811096,
|
| 876 |
+
"learning_rate": 0.0005104975731601208,
|
| 877 |
+
"loss": 3.1196,
|
| 878 |
+
"step": 123
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.006751361161524501,
|
| 882 |
+
"grad_norm": 0.20675747096538544,
|
| 883 |
+
"learning_rate": 0.0005089537547480885,
|
| 884 |
+
"loss": 3.1063,
|
| 885 |
+
"step": 124
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.006805807622504537,
|
| 889 |
+
"grad_norm": 0.20474228262901306,
|
| 890 |
+
"learning_rate": 0.0005074025148547634,
|
| 891 |
+
"loss": 3.1211,
|
| 892 |
+
"step": 125
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.006860254083484573,
|
| 896 |
+
"grad_norm": 0.21194572746753693,
|
| 897 |
+
"learning_rate": 0.0005058440487602918,
|
| 898 |
+
"loss": 3.1456,
|
| 899 |
+
"step": 126
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.00691470054446461,
|
| 903 |
+
"grad_norm": 0.20788611471652985,
|
| 904 |
+
"learning_rate": 0.0005042785526545008,
|
| 905 |
+
"loss": 3.1353,
|
| 906 |
+
"step": 127
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.006969147005444646,
|
| 910 |
+
"grad_norm": 0.19835753738880157,
|
| 911 |
+
"learning_rate": 0.0005027062236122014,
|
| 912 |
+
"loss": 3.0889,
|
| 913 |
+
"step": 128
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.007023593466424683,
|
| 917 |
+
"grad_norm": 0.24613966047763824,
|
| 918 |
+
"learning_rate": 0.0005011272595683787,
|
| 919 |
+
"loss": 3.1023,
|
| 920 |
+
"step": 129
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.0070780399274047185,
|
| 924 |
+
"grad_norm": 0.18778769671916962,
|
| 925 |
+
"learning_rate": 0.000499541859293275,
|
| 926 |
+
"loss": 3.1091,
|
| 927 |
+
"step": 130
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.007132486388384755,
|
| 931 |
+
"grad_norm": 0.22894078493118286,
|
| 932 |
+
"learning_rate": 0.0004979502223673672,
|
| 933 |
+
"loss": 3.0836,
|
| 934 |
+
"step": 131
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.007186932849364791,
|
| 938 |
+
"grad_norm": 0.2452639937400818,
|
| 939 |
+
"learning_rate": 0.0004963525491562421,
|
| 940 |
+
"loss": 3.1285,
|
| 941 |
+
"step": 132
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.007241379310344828,
|
| 945 |
+
"grad_norm": 0.26903706789016724,
|
| 946 |
+
"learning_rate": 0.0004947490407853734,
|
| 947 |
+
"loss": 3.135,
|
| 948 |
+
"step": 133
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.007295825771324864,
|
| 952 |
+
"grad_norm": 0.29524174332618713,
|
| 953 |
+
"learning_rate": 0.0004931398991148025,
|
| 954 |
+
"loss": 3.1204,
|
| 955 |
+
"step": 134
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.0073502722323049,
|
| 959 |
+
"grad_norm": 0.1946076601743698,
|
| 960 |
+
"learning_rate": 0.0004915253267137274,
|
| 961 |
+
"loss": 3.1482,
|
| 962 |
+
"step": 135
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.007404718693284937,
|
| 966 |
+
"grad_norm": 0.3097498416900635,
|
| 967 |
+
"learning_rate": 0.0004899055268350012,
|
| 968 |
+
"loss": 3.1188,
|
| 969 |
+
"step": 136
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.0074591651542649726,
|
| 973 |
+
"grad_norm": 0.23407086730003357,
|
| 974 |
+
"learning_rate": 0.0004882807033895463,
|
| 975 |
+
"loss": 3.1448,
|
| 976 |
+
"step": 137
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.007513611615245009,
|
| 980 |
+
"grad_norm": 0.22323715686798096,
|
| 981 |
+
"learning_rate": 0.0004866510609206841,
|
| 982 |
+
"loss": 3.096,
|
| 983 |
+
"step": 138
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.007568058076225045,
|
| 987 |
+
"grad_norm": 0.23807968199253082,
|
| 988 |
+
"learning_rate": 0.0004850168045783858,
|
| 989 |
+
"loss": 3.1348,
|
| 990 |
+
"step": 139
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.007622504537205082,
|
| 994 |
+
"grad_norm": 0.2817918658256531,
|
| 995 |
+
"learning_rate": 0.0004833781400934471,
|
| 996 |
+
"loss": 3.1215,
|
| 997 |
+
"step": 140
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.007676950998185118,
|
| 1001 |
+
"grad_norm": 0.25021976232528687,
|
| 1002 |
+
"learning_rate": 0.00048173527375158944,
|
| 1003 |
+
"loss": 3.1019,
|
| 1004 |
+
"step": 141
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.007731397459165154,
|
| 1008 |
+
"grad_norm": 0.25827258825302124,
|
| 1009 |
+
"learning_rate": 0.00048008841236749084,
|
| 1010 |
+
"loss": 3.1155,
|
| 1011 |
+
"step": 142
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.007785843920145191,
|
| 1015 |
+
"grad_norm": 0.2191617339849472,
|
| 1016 |
+
"learning_rate": 0.00047843776325875173,
|
| 1017 |
+
"loss": 3.1183,
|
| 1018 |
+
"step": 143
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.007840290381125227,
|
| 1022 |
+
"grad_norm": 0.2649674415588379,
|
| 1023 |
+
"learning_rate": 0.0004767835342197954,
|
| 1024 |
+
"loss": 3.1098,
|
| 1025 |
+
"step": 144
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.007894736842105263,
|
| 1029 |
+
"grad_norm": 0.2287340611219406,
|
| 1030 |
+
"learning_rate": 0.00047512593349571043,
|
| 1031 |
+
"loss": 3.1004,
|
| 1032 |
+
"step": 145
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.0079491833030853,
|
| 1036 |
+
"grad_norm": 0.21950604021549225,
|
| 1037 |
+
"learning_rate": 0.00047346516975603465,
|
| 1038 |
+
"loss": 3.0733,
|
| 1039 |
+
"step": 146
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.008003629764065335,
|
| 1043 |
+
"grad_norm": 0.2002648264169693,
|
| 1044 |
+
"learning_rate": 0.00047180145206848686,
|
| 1045 |
+
"loss": 3.0934,
|
| 1046 |
+
"step": 147
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.008058076225045372,
|
| 1050 |
+
"grad_norm": 0.20886124670505524,
|
| 1051 |
+
"learning_rate": 0.0004701349898726483,
|
| 1052 |
+
"loss": 3.1007,
|
| 1053 |
+
"step": 148
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.008112522686025409,
|
| 1057 |
+
"grad_norm": 0.19583792984485626,
|
| 1058 |
+
"learning_rate": 0.00046846599295359635,
|
| 1059 |
+
"loss": 3.1249,
|
| 1060 |
+
"step": 149
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.008166969147005444,
|
| 1064 |
+
"grad_norm": 0.18620361387729645,
|
| 1065 |
+
"learning_rate": 0.00046679467141549615,
|
| 1066 |
+
"loss": 3.1514,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.00822141560798548,
|
| 1071 |
+
"grad_norm": 0.2266155332326889,
|
| 1072 |
+
"learning_rate": 0.00046512123565515065,
|
| 1073 |
+
"loss": 3.1583,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.008275862068965517,
|
| 1078 |
+
"grad_norm": 0.18331050872802734,
|
| 1079 |
+
"learning_rate": 0.00046344589633551497,
|
| 1080 |
+
"loss": 3.1015,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.008330308529945554,
|
| 1085 |
+
"grad_norm": 0.23021604120731354,
|
| 1086 |
+
"learning_rate": 0.00046176886435917667,
|
| 1087 |
+
"loss": 3.0984,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.00838475499092559,
|
| 1092 |
+
"grad_norm": 0.23210260272026062,
|
| 1093 |
+
"learning_rate": 0.00046009035084180593,
|
| 1094 |
+
"loss": 3.1239,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.008439201451905626,
|
| 1099 |
+
"grad_norm": 0.2083989381790161,
|
| 1100 |
+
"learning_rate": 0.0004584105670855787,
|
| 1101 |
+
"loss": 3.0929,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.008493647912885663,
|
| 1106 |
+
"grad_norm": 0.22709086537361145,
|
| 1107 |
+
"learning_rate": 0.00045672972455257723,
|
| 1108 |
+
"loss": 3.1376,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.008548094373865698,
|
| 1113 |
+
"grad_norm": 0.24739298224449158,
|
| 1114 |
+
"learning_rate": 0.0004550480348381691,
|
| 1115 |
+
"loss": 3.1135,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.008602540834845735,
|
| 1120 |
+
"grad_norm": 0.2940455973148346,
|
| 1121 |
+
"learning_rate": 0.0004533657096443708,
|
| 1122 |
+
"loss": 3.1164,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.008656987295825772,
|
| 1127 |
+
"grad_norm": 0.2183440625667572,
|
| 1128 |
+
"learning_rate": 0.00045168296075319685,
|
| 1129 |
+
"loss": 3.1496,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.008711433756805808,
|
| 1134 |
+
"grad_norm": 0.254537969827652,
|
| 1135 |
+
"learning_rate": 0.00045,
|
| 1136 |
+
"loss": 3.1164,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.008765880217785843,
|
| 1141 |
+
"grad_norm": 0.3475794792175293,
|
| 1142 |
+
"learning_rate": 0.00044831703924680307,
|
| 1143 |
+
"loss": 3.1241,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.00882032667876588,
|
| 1148 |
+
"grad_norm": 0.25378546118736267,
|
| 1149 |
+
"learning_rate": 0.00044663429035562925,
|
| 1150 |
+
"loss": 3.0785,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.008874773139745917,
|
| 1155 |
+
"grad_norm": 0.3237019181251526,
|
| 1156 |
+
"learning_rate": 0.0004449519651618309,
|
| 1157 |
+
"loss": 3.1247,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.008929219600725952,
|
| 1162 |
+
"grad_norm": 0.2086883783340454,
|
| 1163 |
+
"learning_rate": 0.0004432702754474228,
|
| 1164 |
+
"loss": 3.1397,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.008983666061705989,
|
| 1169 |
+
"grad_norm": 0.2567574083805084,
|
| 1170 |
+
"learning_rate": 0.0004415894329144212,
|
| 1171 |
+
"loss": 3.1443,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.009038112522686026,
|
| 1176 |
+
"grad_norm": 0.1955418884754181,
|
| 1177 |
+
"learning_rate": 0.000439909649158194,
|
| 1178 |
+
"loss": 3.1298,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.009092558983666062,
|
| 1183 |
+
"grad_norm": 0.23903286457061768,
|
| 1184 |
+
"learning_rate": 0.00043823113564082325,
|
| 1185 |
+
"loss": 3.1061,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.009147005444646097,
|
| 1190 |
+
"grad_norm": 0.18639686703681946,
|
| 1191 |
+
"learning_rate": 0.00043655410366448495,
|
| 1192 |
+
"loss": 3.1364,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.009201451905626134,
|
| 1197 |
+
"grad_norm": 0.28311777114868164,
|
| 1198 |
+
"learning_rate": 0.0004348787643448493,
|
| 1199 |
+
"loss": 3.1165,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.009255898366606171,
|
| 1204 |
+
"grad_norm": 0.21376436948776245,
|
| 1205 |
+
"learning_rate": 0.0004332053285845038,
|
| 1206 |
+
"loss": 3.1332,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.009310344827586206,
|
| 1211 |
+
"grad_norm": 0.25171706080436707,
|
| 1212 |
+
"learning_rate": 0.0004315340070464036,
|
| 1213 |
+
"loss": 3.0928,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.009364791288566243,
|
| 1218 |
+
"grad_norm": 0.2478170096874237,
|
| 1219 |
+
"learning_rate": 0.0004298650101273517,
|
| 1220 |
+
"loss": 3.1146,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.00941923774954628,
|
| 1225 |
+
"grad_norm": 0.2905559837818146,
|
| 1226 |
+
"learning_rate": 0.0004281985479315131,
|
| 1227 |
+
"loss": 3.1248,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.009473684210526316,
|
| 1232 |
+
"grad_norm": 0.23062513768672943,
|
| 1233 |
+
"learning_rate": 0.00042653483024396527,
|
| 1234 |
+
"loss": 3.0908,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.009528130671506351,
|
| 1239 |
+
"grad_norm": 0.21542707085609436,
|
| 1240 |
+
"learning_rate": 0.0004248740665042895,
|
| 1241 |
+
"loss": 3.116,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.009582577132486388,
|
| 1246 |
+
"grad_norm": 0.2628575563430786,
|
| 1247 |
+
"learning_rate": 0.0004232164657802045,
|
| 1248 |
+
"loss": 3.0974,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.009637023593466425,
|
| 1253 |
+
"grad_norm": 0.2287788689136505,
|
| 1254 |
+
"learning_rate": 0.0004215622367412482,
|
| 1255 |
+
"loss": 3.1229,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.00969147005444646,
|
| 1260 |
+
"grad_norm": 0.23420408368110657,
|
| 1261 |
+
"learning_rate": 0.0004199115876325091,
|
| 1262 |
+
"loss": 3.1408,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.009745916515426497,
|
| 1267 |
+
"grad_norm": 0.19503605365753174,
|
| 1268 |
+
"learning_rate": 0.0004182647262484106,
|
| 1269 |
+
"loss": 3.104,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.009800362976406534,
|
| 1274 |
+
"grad_norm": 0.2603705823421478,
|
| 1275 |
+
"learning_rate": 0.0004166218599065528,
|
| 1276 |
+
"loss": 3.1185,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.00985480943738657,
|
| 1281 |
+
"grad_norm": 0.20003163814544678,
|
| 1282 |
+
"learning_rate": 0.0004149831954216142,
|
| 1283 |
+
"loss": 3.1248,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.009909255898366606,
|
| 1288 |
+
"grad_norm": 0.2520509660243988,
|
| 1289 |
+
"learning_rate": 0.00041334893907931584,
|
| 1290 |
+
"loss": 3.1053,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.009963702359346642,
|
| 1295 |
+
"grad_norm": 0.2309730499982834,
|
| 1296 |
+
"learning_rate": 0.0004117192966104536,
|
| 1297 |
+
"loss": 3.0824,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.01001814882032668,
|
| 1302 |
+
"grad_norm": 0.24436035752296448,
|
| 1303 |
+
"learning_rate": 0.0004100944731649987,
|
| 1304 |
+
"loss": 3.1365,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.010072595281306716,
|
| 1309 |
+
"grad_norm": 0.1982448399066925,
|
| 1310 |
+
"learning_rate": 0.0004084746732862726,
|
| 1311 |
+
"loss": 3.1232,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.010127041742286751,
|
| 1316 |
+
"grad_norm": 0.28294622898101807,
|
| 1317 |
+
"learning_rate": 0.0004068601008851974,
|
| 1318 |
+
"loss": 3.1373,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.010181488203266788,
|
| 1323 |
+
"grad_norm": 0.19457624852657318,
|
| 1324 |
+
"learning_rate": 0.0004052509592146266,
|
| 1325 |
+
"loss": 3.1108,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.010235934664246825,
|
| 1330 |
+
"grad_norm": 0.20812362432479858,
|
| 1331 |
+
"learning_rate": 0.00040364745084375787,
|
| 1332 |
+
"loss": 3.1269,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.01029038112522686,
|
| 1337 |
+
"grad_norm": 0.26139721274375916,
|
| 1338 |
+
"learning_rate": 0.0004020497776326328,
|
| 1339 |
+
"loss": 3.1027,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.010344827586206896,
|
| 1344 |
+
"grad_norm": 0.24709898233413696,
|
| 1345 |
+
"learning_rate": 0.00040045814070672494,
|
| 1346 |
+
"loss": 3.1102,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.010399274047186933,
|
| 1351 |
+
"grad_norm": 0.20621921122074127,
|
| 1352 |
+
"learning_rate": 0.0003988727404316212,
|
| 1353 |
+
"loss": 3.1248,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.01045372050816697,
|
| 1358 |
+
"grad_norm": 0.21720059216022491,
|
| 1359 |
+
"learning_rate": 0.00039729377638779857,
|
| 1360 |
+
"loss": 3.1187,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.010508166969147005,
|
| 1365 |
+
"grad_norm": 0.1787269115447998,
|
| 1366 |
+
"learning_rate": 0.0003957214473454991,
|
| 1367 |
+
"loss": 3.1032,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.010562613430127042,
|
| 1372 |
+
"grad_norm": 0.22008821368217468,
|
| 1373 |
+
"learning_rate": 0.00039415595123970813,
|
| 1374 |
+
"loss": 3.1438,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.010617059891107079,
|
| 1379 |
+
"grad_norm": 0.2438061535358429,
|
| 1380 |
+
"learning_rate": 0.00039259748514523655,
|
| 1381 |
+
"loss": 3.1444,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.010671506352087114,
|
| 1386 |
+
"grad_norm": 0.21327753365039825,
|
| 1387 |
+
"learning_rate": 0.0003910462452519114,
|
| 1388 |
+
"loss": 3.1147,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.01072595281306715,
|
| 1393 |
+
"grad_norm": 0.22642219066619873,
|
| 1394 |
+
"learning_rate": 0.0003895024268398792,
|
| 1395 |
+
"loss": 3.1221,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.010780399274047187,
|
| 1400 |
+
"grad_norm": 0.1771782636642456,
|
| 1401 |
+
"learning_rate": 0.00038796622425502195,
|
| 1402 |
+
"loss": 3.1153,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.010834845735027224,
|
| 1407 |
+
"grad_norm": 0.25223565101623535,
|
| 1408 |
+
"learning_rate": 0.00038643783088449163,
|
| 1409 |
+
"loss": 3.1249,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.010889292196007259,
|
| 1414 |
+
"grad_norm": 0.15892428159713745,
|
| 1415 |
+
"learning_rate": 0.00038491743913236624,
|
| 1416 |
+
"loss": 3.1273,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.010889292196007259,
|
| 1421 |
+
"eval_loss": 3.0130114555358887,
|
| 1422 |
+
"eval_runtime": 69.209,
|
| 1423 |
+
"eval_samples_per_second": 62.347,
|
| 1424 |
+
"eval_steps_per_second": 15.59,
|
| 1425 |
+
"step": 200
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 0.010943738656987296,
|
| 1429 |
+
"grad_norm": 0.2665696144104004,
|
| 1430 |
+
"learning_rate": 0.0003834052403954277,
|
| 1431 |
+
"loss": 3.1162,
|
| 1432 |
+
"step": 201
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 0.010998185117967333,
|
| 1436 |
+
"grad_norm": 0.2027243971824646,
|
| 1437 |
+
"learning_rate": 0.00038190142503906794,
|
| 1438 |
+
"loss": 3.0788,
|
| 1439 |
+
"step": 202
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 0.011052631578947368,
|
| 1443 |
+
"grad_norm": 0.2640676498413086,
|
| 1444 |
+
"learning_rate": 0.00038040618237332485,
|
| 1445 |
+
"loss": 3.0978,
|
| 1446 |
+
"step": 203
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 0.011107078039927405,
|
| 1450 |
+
"grad_norm": 0.21408383548259735,
|
| 1451 |
+
"learning_rate": 0.0003789197006290502,
|
| 1452 |
+
"loss": 3.1155,
|
| 1453 |
+
"step": 204
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 0.011161524500907441,
|
| 1457 |
+
"grad_norm": 0.21028189361095428,
|
| 1458 |
+
"learning_rate": 0.00037744216693421403,
|
| 1459 |
+
"loss": 3.0896,
|
| 1460 |
+
"step": 205
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.011215970961887478,
|
| 1464 |
+
"grad_norm": 0.183872789144516,
|
| 1465 |
+
"learning_rate": 0.00037597376729034794,
|
| 1466 |
+
"loss": 3.1179,
|
| 1467 |
+
"step": 206
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.011270417422867513,
|
| 1471 |
+
"grad_norm": 0.24802212417125702,
|
| 1472 |
+
"learning_rate": 0.00037451468654912994,
|
| 1473 |
+
"loss": 3.1283,
|
| 1474 |
+
"step": 207
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 0.01132486388384755,
|
| 1478 |
+
"grad_norm": 0.20964276790618896,
|
| 1479 |
+
"learning_rate": 0.00037306510838911404,
|
| 1480 |
+
"loss": 3.0941,
|
| 1481 |
+
"step": 208
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.011379310344827587,
|
| 1485 |
+
"grad_norm": 0.2240600436925888,
|
| 1486 |
+
"learning_rate": 0.00037162521529260763,
|
| 1487 |
+
"loss": 3.1226,
|
| 1488 |
+
"step": 209
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 0.011433756805807622,
|
| 1492 |
+
"grad_norm": 0.23930035531520844,
|
| 1493 |
+
"learning_rate": 0.00037019518852269954,
|
| 1494 |
+
"loss": 3.111,
|
| 1495 |
+
"step": 210
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 0.011488203266787659,
|
| 1499 |
+
"grad_norm": 0.21928225457668304,
|
| 1500 |
+
"learning_rate": 0.0003687752081004411,
|
| 1501 |
+
"loss": 3.1014,
|
| 1502 |
+
"step": 211
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 0.011542649727767695,
|
| 1506 |
+
"grad_norm": 0.24399928748607635,
|
| 1507 |
+
"learning_rate": 0.0003673654527821846,
|
| 1508 |
+
"loss": 3.1345,
|
| 1509 |
+
"step": 212
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 0.011597096188747732,
|
| 1513 |
+
"grad_norm": 0.23558929562568665,
|
| 1514 |
+
"learning_rate": 0.00036596610003707954,
|
| 1515 |
+
"loss": 3.0852,
|
| 1516 |
+
"step": 213
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 0.011651542649727767,
|
| 1520 |
+
"grad_norm": 0.19627036154270172,
|
| 1521 |
+
"learning_rate": 0.00036457732602473216,
|
| 1522 |
+
"loss": 3.1098,
|
| 1523 |
+
"step": 214
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 0.011705989110707804,
|
| 1527 |
+
"grad_norm": 0.26581332087516785,
|
| 1528 |
+
"learning_rate": 0.0003631993055730291,
|
| 1529 |
+
"loss": 3.1467,
|
| 1530 |
+
"step": 215
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.01176043557168784,
|
| 1534 |
+
"grad_norm": 0.2310531884431839,
|
| 1535 |
+
"learning_rate": 0.000361832212156129,
|
| 1536 |
+
"loss": 3.1242,
|
| 1537 |
+
"step": 216
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 0.011814882032667876,
|
| 1541 |
+
"grad_norm": 0.2427210807800293,
|
| 1542 |
+
"learning_rate": 0.00036047621787262444,
|
| 1543 |
+
"loss": 3.1144,
|
| 1544 |
+
"step": 217
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 0.011869328493647913,
|
| 1548 |
+
"grad_norm": 0.2132464200258255,
|
| 1549 |
+
"learning_rate": 0.000359131493423877,
|
| 1550 |
+
"loss": 3.1184,
|
| 1551 |
+
"step": 218
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 0.01192377495462795,
|
| 1555 |
+
"grad_norm": 0.2877713441848755,
|
| 1556 |
+
"learning_rate": 0.0003577982080925284,
|
| 1557 |
+
"loss": 3.0916,
|
| 1558 |
+
"step": 219
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 0.011978221415607986,
|
| 1562 |
+
"grad_norm": 0.20918139815330505,
|
| 1563 |
+
"learning_rate": 0.00035647652972119,
|
| 1564 |
+
"loss": 3.1299,
|
| 1565 |
+
"step": 220
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 0.012032667876588021,
|
| 1569 |
+
"grad_norm": 0.23674660921096802,
|
| 1570 |
+
"learning_rate": 0.00035516662469131356,
|
| 1571 |
+
"loss": 3.1292,
|
| 1572 |
+
"step": 221
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 0.012087114337568058,
|
| 1576 |
+
"grad_norm": 0.1943867802619934,
|
| 1577 |
+
"learning_rate": 0.0003538686579022464,
|
| 1578 |
+
"loss": 3.1258,
|
| 1579 |
+
"step": 222
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 0.012141560798548095,
|
| 1583 |
+
"grad_norm": 0.34807291626930237,
|
| 1584 |
+
"learning_rate": 0.00035258279275047246,
|
| 1585 |
+
"loss": 3.1328,
|
| 1586 |
+
"step": 223
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.01219600725952813,
|
| 1590 |
+
"grad_norm": 0.20535360276699066,
|
| 1591 |
+
"learning_rate": 0.0003513091911090431,
|
| 1592 |
+
"loss": 3.1184,
|
| 1593 |
+
"step": 224
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 0.012250453720508167,
|
| 1597 |
+
"grad_norm": 0.26084408164024353,
|
| 1598 |
+
"learning_rate": 0.00035004801330719936,
|
| 1599 |
+
"loss": 3.1014,
|
| 1600 |
+
"step": 225
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.012304900181488203,
|
| 1604 |
+
"grad_norm": 0.2201356142759323,
|
| 1605 |
+
"learning_rate": 0.0003487994181101888,
|
| 1606 |
+
"loss": 3.0941,
|
| 1607 |
+
"step": 226
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 0.01235934664246824,
|
| 1611 |
+
"grad_norm": 0.20090270042419434,
|
| 1612 |
+
"learning_rate": 0.00034756356269927894,
|
| 1613 |
+
"loss": 3.093,
|
| 1614 |
+
"step": 227
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.012413793103448275,
|
| 1618 |
+
"grad_norm": 0.22596506774425507,
|
| 1619 |
+
"learning_rate": 0.00034634060265197026,
|
| 1620 |
+
"loss": 3.1011,
|
| 1621 |
+
"step": 228
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 0.012468239564428312,
|
| 1625 |
+
"grad_norm": 0.2359432578086853,
|
| 1626 |
+
"learning_rate": 0.00034513069192241137,
|
| 1627 |
+
"loss": 3.1118,
|
| 1628 |
+
"step": 229
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 0.012522686025408349,
|
| 1632 |
+
"grad_norm": 0.2078094184398651,
|
| 1633 |
+
"learning_rate": 0.0003439339828220179,
|
| 1634 |
+
"loss": 3.0898,
|
| 1635 |
+
"step": 230
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 0.012577132486388384,
|
| 1639 |
+
"grad_norm": 0.19436654448509216,
|
| 1640 |
+
"learning_rate": 0.00034275062600029865,
|
| 1641 |
+
"loss": 3.0997,
|
| 1642 |
+
"step": 231
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 0.01263157894736842,
|
| 1646 |
+
"grad_norm": 0.1749400496482849,
|
| 1647 |
+
"learning_rate": 0.0003415807704258913,
|
| 1648 |
+
"loss": 3.1013,
|
| 1649 |
+
"step": 232
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 0.012686025408348458,
|
| 1653 |
+
"grad_norm": 0.31390950083732605,
|
| 1654 |
+
"learning_rate": 0.00034042456336780833,
|
| 1655 |
+
"loss": 3.1088,
|
| 1656 |
+
"step": 233
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 0.012740471869328494,
|
| 1660 |
+
"grad_norm": 0.21334995329380035,
|
| 1661 |
+
"learning_rate": 0.00033928215037689886,
|
| 1662 |
+
"loss": 3.1367,
|
| 1663 |
+
"step": 234
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 0.01279491833030853,
|
| 1667 |
+
"grad_norm": 0.2765377461910248,
|
| 1668 |
+
"learning_rate": 0.00033815367526752516,
|
| 1669 |
+
"loss": 3.0837,
|
| 1670 |
+
"step": 235
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.012849364791288566,
|
| 1674 |
+
"grad_norm": 0.21292102336883545,
|
| 1675 |
+
"learning_rate": 0.0003370392800994583,
|
| 1676 |
+
"loss": 3.116,
|
| 1677 |
+
"step": 236
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 0.012903811252268603,
|
| 1681 |
+
"grad_norm": 0.22039784491062164,
|
| 1682 |
+
"learning_rate": 0.0003359391051599953,
|
| 1683 |
+
"loss": 3.1287,
|
| 1684 |
+
"step": 237
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 0.012958257713248638,
|
| 1688 |
+
"grad_norm": 0.31274768710136414,
|
| 1689 |
+
"learning_rate": 0.0003348532889462983,
|
| 1690 |
+
"loss": 3.103,
|
| 1691 |
+
"step": 238
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 0.013012704174228675,
|
| 1695 |
+
"grad_norm": 0.2530403435230255,
|
| 1696 |
+
"learning_rate": 0.00033378196814795987,
|
| 1697 |
+
"loss": 3.1088,
|
| 1698 |
+
"step": 239
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 0.013067150635208712,
|
| 1702 |
+
"grad_norm": 0.25343602895736694,
|
| 1703 |
+
"learning_rate": 0.0003327252776297955,
|
| 1704 |
+
"loss": 3.0815,
|
| 1705 |
+
"step": 240
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.013121597096188748,
|
| 1709 |
+
"grad_norm": 0.2652718722820282,
|
| 1710 |
+
"learning_rate": 0.0003316833504148663,
|
| 1711 |
+
"loss": 3.1188,
|
| 1712 |
+
"step": 241
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.013176043557168783,
|
| 1716 |
+
"grad_norm": 0.2119700163602829,
|
| 1717 |
+
"learning_rate": 0.0003306563176677328,
|
| 1718 |
+
"loss": 3.1262,
|
| 1719 |
+
"step": 242
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 0.01323049001814882,
|
| 1723 |
+
"grad_norm": 0.2630724310874939,
|
| 1724 |
+
"learning_rate": 0.00032964430867794326,
|
| 1725 |
+
"loss": 3.1183,
|
| 1726 |
+
"step": 243
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 0.013284936479128857,
|
| 1730 |
+
"grad_norm": 0.2809687554836273,
|
| 1731 |
+
"learning_rate": 0.00032864745084375783,
|
| 1732 |
+
"loss": 3.1087,
|
| 1733 |
+
"step": 244
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 0.013339382940108892,
|
| 1737 |
+
"grad_norm": 0.24847863614559174,
|
| 1738 |
+
"learning_rate": 0.00032766586965611095,
|
| 1739 |
+
"loss": 3.1053,
|
| 1740 |
+
"step": 245
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 0.013393829401088929,
|
| 1744 |
+
"grad_norm": 0.2698868215084076,
|
| 1745 |
+
"learning_rate": 0.00032669968868281353,
|
| 1746 |
+
"loss": 3.0977,
|
| 1747 |
+
"step": 246
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 0.013448275862068966,
|
| 1751 |
+
"grad_norm": 0.33301204442977905,
|
| 1752 |
+
"learning_rate": 0.0003257490295529975,
|
| 1753 |
+
"loss": 3.098,
|
| 1754 |
+
"step": 247
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.013502722323049002,
|
| 1758 |
+
"grad_norm": 0.2521134912967682,
|
| 1759 |
+
"learning_rate": 0.0003248140119418046,
|
| 1760 |
+
"loss": 3.0924,
|
| 1761 |
+
"step": 248
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 0.013557168784029038,
|
| 1765 |
+
"grad_norm": 0.22425313293933868,
|
| 1766 |
+
"learning_rate": 0.00032389475355532044,
|
| 1767 |
+
"loss": 3.0911,
|
| 1768 |
+
"step": 249
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 0.013611615245009074,
|
| 1772 |
+
"grad_norm": 0.2841155230998993,
|
| 1773 |
+
"learning_rate": 0.00032299137011575734,
|
| 1774 |
+
"loss": 3.1019,
|
| 1775 |
+
"step": 250
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 0.013666061705989111,
|
| 1779 |
+
"grad_norm": 0.19375889003276825,
|
| 1780 |
+
"learning_rate": 0.00032210397534688617,
|
| 1781 |
+
"loss": 3.1183,
|
| 1782 |
+
"step": 251
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 0.013720508166969146,
|
| 1786 |
+
"grad_norm": 0.26036033034324646,
|
| 1787 |
+
"learning_rate": 0.00032123268095972005,
|
| 1788 |
+
"loss": 3.1296,
|
| 1789 |
+
"step": 252
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 0.013774954627949183,
|
| 1793 |
+
"grad_norm": 0.24063162505626678,
|
| 1794 |
+
"learning_rate": 0.0003203775966384518,
|
| 1795 |
+
"loss": 3.142,
|
| 1796 |
+
"step": 253
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 0.01382940108892922,
|
| 1800 |
+
"grad_norm": 0.30820685625076294,
|
| 1801 |
+
"learning_rate": 0.0003195388300266457,
|
| 1802 |
+
"loss": 3.0998,
|
| 1803 |
+
"step": 254
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 0.013883847549909257,
|
| 1807 |
+
"grad_norm": 0.18717393279075623,
|
| 1808 |
+
"learning_rate": 0.0003187164867136869,
|
| 1809 |
+
"loss": 3.1363,
|
| 1810 |
+
"step": 255
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.013938294010889292,
|
| 1814 |
+
"grad_norm": 0.22616641223430634,
|
| 1815 |
+
"learning_rate": 0.0003179106702214893,
|
| 1816 |
+
"loss": 3.1284,
|
| 1817 |
+
"step": 256
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 0.013992740471869328,
|
| 1821 |
+
"grad_norm": 0.19695289433002472,
|
| 1822 |
+
"learning_rate": 0.000317121481991463,
|
| 1823 |
+
"loss": 3.0859,
|
| 1824 |
+
"step": 257
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.014047186932849365,
|
| 1828 |
+
"grad_norm": 0.24756500124931335,
|
| 1829 |
+
"learning_rate": 0.0003163490213717448,
|
| 1830 |
+
"loss": 3.1187,
|
| 1831 |
+
"step": 258
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 0.014101633393829402,
|
| 1835 |
+
"grad_norm": 0.2162931114435196,
|
| 1836 |
+
"learning_rate": 0.00031559338560469116,
|
| 1837 |
+
"loss": 3.1066,
|
| 1838 |
+
"step": 259
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.014156079854809437,
|
| 1842 |
+
"grad_norm": 0.2389383465051651,
|
| 1843 |
+
"learning_rate": 0.0003148546698146371,
|
| 1844 |
+
"loss": 3.1096,
|
| 1845 |
+
"step": 260
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 0.014210526315789474,
|
| 1849 |
+
"grad_norm": 0.1939517855644226,
|
| 1850 |
+
"learning_rate": 0.0003141329669959209,
|
| 1851 |
+
"loss": 3.115,
|
| 1852 |
+
"step": 261
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 0.01426497277676951,
|
| 1856 |
+
"grad_norm": 0.24441573023796082,
|
| 1857 |
+
"learning_rate": 0.00031342836800117763,
|
| 1858 |
+
"loss": 3.0954,
|
| 1859 |
+
"step": 262
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 0.014319419237749546,
|
| 1863 |
+
"grad_norm": 0.31514814496040344,
|
| 1864 |
+
"learning_rate": 0.000312740961529902,
|
| 1865 |
+
"loss": 3.1259,
|
| 1866 |
+
"step": 263
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 0.014373865698729582,
|
| 1870 |
+
"grad_norm": 0.28324607014656067,
|
| 1871 |
+
"learning_rate": 0.00031207083411728236,
|
| 1872 |
+
"loss": 3.1088,
|
| 1873 |
+
"step": 264
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.01442831215970962,
|
| 1877 |
+
"grad_norm": 0.21894101798534393,
|
| 1878 |
+
"learning_rate": 0.00031141807012330695,
|
| 1879 |
+
"loss": 3.1071,
|
| 1880 |
+
"step": 265
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 0.014482758620689656,
|
| 1884 |
+
"grad_norm": 0.32468879222869873,
|
| 1885 |
+
"learning_rate": 0.0003107827517221441,
|
| 1886 |
+
"loss": 3.1159,
|
| 1887 |
+
"step": 266
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 0.014537205081669691,
|
| 1891 |
+
"grad_norm": 0.2448977530002594,
|
| 1892 |
+
"learning_rate": 0.00031016495889179787,
|
| 1893 |
+
"loss": 3.1091,
|
| 1894 |
+
"step": 267
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 0.014591651542649728,
|
| 1898 |
+
"grad_norm": 0.3004207909107208,
|
| 1899 |
+
"learning_rate": 0.0003095647694040394,
|
| 1900 |
+
"loss": 3.1091,
|
| 1901 |
+
"step": 268
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 0.014646098003629765,
|
| 1905 |
+
"grad_norm": 0.2677028179168701,
|
| 1906 |
+
"learning_rate": 0.0003089822588146171,
|
| 1907 |
+
"loss": 3.1069,
|
| 1908 |
+
"step": 269
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 0.0147005444646098,
|
| 1912 |
+
"grad_norm": 0.26184970140457153,
|
| 1913 |
+
"learning_rate": 0.0003084175004537448,
|
| 1914 |
+
"loss": 3.1272,
|
| 1915 |
+
"step": 270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 0.014754990925589836,
|
| 1919 |
+
"grad_norm": 0.2534137964248657,
|
| 1920 |
+
"learning_rate": 0.0003078705654168706,
|
| 1921 |
+
"loss": 3.0968,
|
| 1922 |
+
"step": 271
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 0.014809437386569873,
|
| 1926 |
+
"grad_norm": 0.34196603298187256,
|
| 1927 |
+
"learning_rate": 0.0003073415225557269,
|
| 1928 |
+
"loss": 3.0834,
|
| 1929 |
+
"step": 272
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 0.01486388384754991,
|
| 1933 |
+
"grad_norm": 0.20533894002437592,
|
| 1934 |
+
"learning_rate": 0.0003068304384696629,
|
| 1935 |
+
"loss": 3.11,
|
| 1936 |
+
"step": 273
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 0.014918330308529945,
|
| 1940 |
+
"grad_norm": 0.2631680369377136,
|
| 1941 |
+
"learning_rate": 0.00030633737749726045,
|
| 1942 |
+
"loss": 3.0984,
|
| 1943 |
+
"step": 274
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 0.014972776769509982,
|
| 1947 |
+
"grad_norm": 0.24283327162265778,
|
| 1948 |
+
"learning_rate": 0.0003058624017082351,
|
| 1949 |
+
"loss": 3.1158,
|
| 1950 |
+
"step": 275
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 0.015027223230490019,
|
| 1954 |
+
"grad_norm": 0.23213128745555878,
|
| 1955 |
+
"learning_rate": 0.000305405570895622,
|
| 1956 |
+
"loss": 3.1186,
|
| 1957 |
+
"step": 276
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 0.015081669691470054,
|
| 1961 |
+
"grad_norm": 0.2027149498462677,
|
| 1962 |
+
"learning_rate": 0.00030496694256824903,
|
| 1963 |
+
"loss": 3.0893,
|
| 1964 |
+
"step": 277
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 0.01513611615245009,
|
| 1968 |
+
"grad_norm": 0.3051553964614868,
|
| 1969 |
+
"learning_rate": 0.00030454657194349695,
|
| 1970 |
+
"loss": 3.1062,
|
| 1971 |
+
"step": 278
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 0.015190562613430127,
|
| 1975 |
+
"grad_norm": 0.28350913524627686,
|
| 1976 |
+
"learning_rate": 0.00030414451194034846,
|
| 1977 |
+
"loss": 3.1115,
|
| 1978 |
+
"step": 279
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 0.015245009074410164,
|
| 1982 |
+
"grad_norm": 0.28886678814888,
|
| 1983 |
+
"learning_rate": 0.00030376081317272645,
|
| 1984 |
+
"loss": 3.1113,
|
| 1985 |
+
"step": 280
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 0.0152994555353902,
|
| 1989 |
+
"grad_norm": 0.24702845513820648,
|
| 1990 |
+
"learning_rate": 0.0003033955239431221,
|
| 1991 |
+
"loss": 3.092,
|
| 1992 |
+
"step": 281
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 0.015353901996370236,
|
| 1996 |
+
"grad_norm": 0.20456074178218842,
|
| 1997 |
+
"learning_rate": 0.00030304869023651464,
|
| 1998 |
+
"loss": 3.1087,
|
| 1999 |
+
"step": 282
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 0.015408348457350273,
|
| 2003 |
+
"grad_norm": 0.19598916172981262,
|
| 2004 |
+
"learning_rate": 0.0003027203557145822,
|
| 2005 |
+
"loss": 3.1357,
|
| 2006 |
+
"step": 283
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 0.015462794918330308,
|
| 2010 |
+
"grad_norm": 0.33903974294662476,
|
| 2011 |
+
"learning_rate": 0.0003024105617102055,
|
| 2012 |
+
"loss": 3.1097,
|
| 2013 |
+
"step": 284
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 0.015517241379310345,
|
| 2017 |
+
"grad_norm": 0.22449573874473572,
|
| 2018 |
+
"learning_rate": 0.0003021193472222646,
|
| 2019 |
+
"loss": 3.1204,
|
| 2020 |
+
"step": 285
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 0.015571687840290381,
|
| 2024 |
+
"grad_norm": 0.28471046686172485,
|
| 2025 |
+
"learning_rate": 0.0003018467489107293,
|
| 2026 |
+
"loss": 3.115,
|
| 2027 |
+
"step": 286
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 0.015626134301270418,
|
| 2031 |
+
"grad_norm": 0.2602684497833252,
|
| 2032 |
+
"learning_rate": 0.0003015928010920444,
|
| 2033 |
+
"loss": 3.1306,
|
| 2034 |
+
"step": 287
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 0.015680580762250453,
|
| 2038 |
+
"grad_norm": 0.23596693575382233,
|
| 2039 |
+
"learning_rate": 0.0003013575357348098,
|
| 2040 |
+
"loss": 3.1111,
|
| 2041 |
+
"step": 288
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 0.01573502722323049,
|
| 2045 |
+
"grad_norm": 0.3110821843147278,
|
| 2046 |
+
"learning_rate": 0.0003011409824557554,
|
| 2047 |
+
"loss": 3.0869,
|
| 2048 |
+
"step": 289
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 0.015789473684210527,
|
| 2052 |
+
"grad_norm": 0.22817878425121307,
|
| 2053 |
+
"learning_rate": 0.00030094316851601356,
|
| 2054 |
+
"loss": 3.1123,
|
| 2055 |
+
"step": 290
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 0.015843920145190562,
|
| 2059 |
+
"grad_norm": 0.26526954770088196,
|
| 2060 |
+
"learning_rate": 0.00030076411881768716,
|
| 2061 |
+
"loss": 3.1038,
|
| 2062 |
+
"step": 291
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 0.0158983666061706,
|
| 2066 |
+
"grad_norm": 0.24120832979679108,
|
| 2067 |
+
"learning_rate": 0.0003006038559007141,
|
| 2068 |
+
"loss": 3.0763,
|
| 2069 |
+
"step": 292
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 0.015952813067150635,
|
| 2073 |
+
"grad_norm": 0.2750589847564697,
|
| 2074 |
+
"learning_rate": 0.0003004623999400308,
|
| 2075 |
+
"loss": 3.1197,
|
| 2076 |
+
"step": 293
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 0.01600725952813067,
|
| 2080 |
+
"grad_norm": 0.26484569907188416,
|
| 2081 |
+
"learning_rate": 0.0003003397687430316,
|
| 2082 |
+
"loss": 3.0945,
|
| 2083 |
+
"step": 294
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 0.01606170598911071,
|
| 2087 |
+
"grad_norm": 0.22410178184509277,
|
| 2088 |
+
"learning_rate": 0.0003002359777473275,
|
| 2089 |
+
"loss": 3.1024,
|
| 2090 |
+
"step": 295
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 0.016116152450090744,
|
| 2094 |
+
"grad_norm": 0.23759888112545013,
|
| 2095 |
+
"learning_rate": 0.00030015104001880274,
|
| 2096 |
+
"loss": 3.1167,
|
| 2097 |
+
"step": 296
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 0.01617059891107078,
|
| 2101 |
+
"grad_norm": 0.21586690843105316,
|
| 2102 |
+
"learning_rate": 0.00030008496624996995,
|
| 2103 |
+
"loss": 3.0945,
|
| 2104 |
+
"step": 297
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 0.016225045372050818,
|
| 2108 |
+
"grad_norm": 0.23178769648075104,
|
| 2109 |
+
"learning_rate": 0.00030003776475862396,
|
| 2110 |
+
"loss": 3.095,
|
| 2111 |
+
"step": 298
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 0.016279491833030853,
|
| 2115 |
+
"grad_norm": 0.24513190984725952,
|
| 2116 |
+
"learning_rate": 0.0003000094414867949,
|
| 2117 |
+
"loss": 3.0728,
|
| 2118 |
+
"step": 299
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 0.016333938294010888,
|
| 2122 |
+
"grad_norm": 0.20249220728874207,
|
| 2123 |
+
"learning_rate": 0.0003,
|
| 2124 |
+
"loss": 3.0881,
|
| 2125 |
+
"step": 300
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 0.016333938294010888,
|
| 2129 |
+
"eval_loss": 3.00935697555542,
|
| 2130 |
+
"eval_runtime": 69.1435,
|
| 2131 |
+
"eval_samples_per_second": 62.406,
|
| 2132 |
+
"eval_steps_per_second": 15.605,
|
| 2133 |
+
"step": 300
|
| 2134 |
+
}
|
| 2135 |
+
],
|
| 2136 |
+
"logging_steps": 1,
|
| 2137 |
+
"max_steps": 300,
|
| 2138 |
+
"num_input_tokens_seen": 0,
|
| 2139 |
+
"num_train_epochs": 1,
|
| 2140 |
+
"save_steps": 100,
|
| 2141 |
+
"stateful_callbacks": {
|
| 2142 |
+
"TrainerControl": {
|
| 2143 |
+
"args": {
|
| 2144 |
+
"should_epoch_stop": false,
|
| 2145 |
+
"should_evaluate": false,
|
| 2146 |
+
"should_log": false,
|
| 2147 |
+
"should_save": true,
|
| 2148 |
+
"should_training_stop": true
|
| 2149 |
+
},
|
| 2150 |
+
"attributes": {}
|
| 2151 |
+
}
|
| 2152 |
+
},
|
| 2153 |
+
"total_flos": 7.5252105216e+16,
|
| 2154 |
+
"train_batch_size": 40,
|
| 2155 |
+
"trial_name": null,
|
| 2156 |
+
"trial_params": null
|
| 2157 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:235f25ca5df2a2718967b5215c20a03890b09a758951caf17f4b82bf949b1030
|
| 3 |
+
size 5304
|