Upload folder using huggingface_hub
Browse files- config.json +38 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +21 -0
- trainer_state.json +2456 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_ctx": 1024,
|
| 14 |
+
"n_embd": 768,
|
| 15 |
+
"n_head": 12,
|
| 16 |
+
"n_inner": null,
|
| 17 |
+
"n_layer": 12,
|
| 18 |
+
"n_positions": 1024,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"task_specific_params": {
|
| 29 |
+
"text-generation": {
|
| 30 |
+
"do_sample": true,
|
| 31 |
+
"max_length": 50
|
| 32 |
+
}
|
| 33 |
+
},
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.50.3",
|
| 36 |
+
"use_cache": true,
|
| 37 |
+
"vocab_size": 50257
|
| 38 |
+
}
|
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": "4.50.3"
|
| 6 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f86fad2c35c2522dd02a15785b52710700b16350e654b6f19c76c1b3c08f956b
|
| 3 |
+
size 497774208
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d652a7f3d0e63b6742f89114778512d67a4e7e8def7bc82517c9262428dce326
|
| 3 |
+
size 995642298
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:718e95ddd6d3415c381350c7da651540b210fe43cc2d5a69bb3aea0e2e5a5f88
|
| 3 |
+
size 14244
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d82b9dd776962ec88ad6eb8cc76d7c238e39e374f1fbca77da53cf804f03119
|
| 3 |
+
size 988
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2df12dfb8a4862243ffba226c514e40fd121acaa1deaddacb9e79a51fde9997b
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
| 6 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
}
|
| 12 |
+
},
|
| 13 |
+
"bos_token": "<|endoftext|>",
|
| 14 |
+
"clean_up_tokenization_spaces": false,
|
| 15 |
+
"eos_token": "<|endoftext|>",
|
| 16 |
+
"extra_special_tokens": {},
|
| 17 |
+
"model_max_length": 1024,
|
| 18 |
+
"pad_token": "<|endoftext|>",
|
| 19 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 20 |
+
"unk_token": "<|endoftext|>"
|
| 21 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2456 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3468,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00865051903114187,
|
| 14 |
+
"grad_norm": 19.23911476135254,
|
| 15 |
+
"learning_rate": 4.9913494809688585e-05,
|
| 16 |
+
"loss": 3.5987,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01730103806228374,
|
| 21 |
+
"grad_norm": 38.634117126464844,
|
| 22 |
+
"learning_rate": 4.9769319492502886e-05,
|
| 23 |
+
"loss": 3.3494,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.025951557093425604,
|
| 28 |
+
"grad_norm": 16.30203628540039,
|
| 29 |
+
"learning_rate": 4.962514417531719e-05,
|
| 30 |
+
"loss": 2.7171,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03460207612456748,
|
| 35 |
+
"grad_norm": 0.0,
|
| 36 |
+
"learning_rate": 4.949538638985006e-05,
|
| 37 |
+
"loss": 3.1395,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04325259515570934,
|
| 42 |
+
"grad_norm": 19.528146743774414,
|
| 43 |
+
"learning_rate": 4.935121107266436e-05,
|
| 44 |
+
"loss": 3.1388,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05190311418685121,
|
| 49 |
+
"grad_norm": 44.69698715209961,
|
| 50 |
+
"learning_rate": 4.920703575547866e-05,
|
| 51 |
+
"loss": 3.1337,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06055363321799308,
|
| 56 |
+
"grad_norm": 19.81340980529785,
|
| 57 |
+
"learning_rate": 4.9077277970011535e-05,
|
| 58 |
+
"loss": 3.2477,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06920415224913495,
|
| 63 |
+
"grad_norm": 11.779070854187012,
|
| 64 |
+
"learning_rate": 4.893310265282584e-05,
|
| 65 |
+
"loss": 3.1636,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07785467128027682,
|
| 70 |
+
"grad_norm": 15.74388599395752,
|
| 71 |
+
"learning_rate": 4.8788927335640136e-05,
|
| 72 |
+
"loss": 3.0162,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08650519031141868,
|
| 77 |
+
"grad_norm": 18.48145866394043,
|
| 78 |
+
"learning_rate": 4.864475201845444e-05,
|
| 79 |
+
"loss": 3.1576,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09515570934256055,
|
| 84 |
+
"grad_norm": 13.73051643371582,
|
| 85 |
+
"learning_rate": 4.8500576701268744e-05,
|
| 86 |
+
"loss": 2.8694,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10380622837370242,
|
| 91 |
+
"grad_norm": 14.890731811523438,
|
| 92 |
+
"learning_rate": 4.835640138408305e-05,
|
| 93 |
+
"loss": 3.267,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.11245674740484429,
|
| 98 |
+
"grad_norm": 15.936467170715332,
|
| 99 |
+
"learning_rate": 4.821222606689735e-05,
|
| 100 |
+
"loss": 2.827,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.12110726643598616,
|
| 105 |
+
"grad_norm": 16.747299194335938,
|
| 106 |
+
"learning_rate": 4.806805074971165e-05,
|
| 107 |
+
"loss": 3.0362,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.12975778546712802,
|
| 112 |
+
"grad_norm": 19.833837509155273,
|
| 113 |
+
"learning_rate": 4.792387543252596e-05,
|
| 114 |
+
"loss": 3.0446,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.1384083044982699,
|
| 119 |
+
"grad_norm": 21.07550048828125,
|
| 120 |
+
"learning_rate": 4.777970011534025e-05,
|
| 121 |
+
"loss": 2.9943,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.14705882352941177,
|
| 126 |
+
"grad_norm": 0.0,
|
| 127 |
+
"learning_rate": 4.763552479815456e-05,
|
| 128 |
+
"loss": 2.9449,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.15570934256055363,
|
| 133 |
+
"grad_norm": 25.829200744628906,
|
| 134 |
+
"learning_rate": 4.749134948096886e-05,
|
| 135 |
+
"loss": 3.1063,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1643598615916955,
|
| 140 |
+
"grad_norm": 16.745203018188477,
|
| 141 |
+
"learning_rate": 4.734717416378317e-05,
|
| 142 |
+
"loss": 3.0687,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.17301038062283736,
|
| 147 |
+
"grad_norm": 19.552099227905273,
|
| 148 |
+
"learning_rate": 4.720299884659746e-05,
|
| 149 |
+
"loss": 3.0418,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.18166089965397925,
|
| 154 |
+
"grad_norm": 24.787086486816406,
|
| 155 |
+
"learning_rate": 4.705882352941177e-05,
|
| 156 |
+
"loss": 3.3645,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.1903114186851211,
|
| 161 |
+
"grad_norm": 22.9111385345459,
|
| 162 |
+
"learning_rate": 4.691464821222607e-05,
|
| 163 |
+
"loss": 2.7289,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.19896193771626297,
|
| 168 |
+
"grad_norm": 0.0,
|
| 169 |
+
"learning_rate": 4.6770472895040375e-05,
|
| 170 |
+
"loss": 3.1899,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.20761245674740483,
|
| 175 |
+
"grad_norm": 17.025423049926758,
|
| 176 |
+
"learning_rate": 4.6626297577854675e-05,
|
| 177 |
+
"loss": 3.0219,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.21626297577854672,
|
| 182 |
+
"grad_norm": 23.269264221191406,
|
| 183 |
+
"learning_rate": 4.6482122260668976e-05,
|
| 184 |
+
"loss": 3.0619,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.22491349480968859,
|
| 189 |
+
"grad_norm": 12.333551406860352,
|
| 190 |
+
"learning_rate": 4.6337946943483276e-05,
|
| 191 |
+
"loss": 3.0275,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.23356401384083045,
|
| 196 |
+
"grad_norm": 9.938949584960938,
|
| 197 |
+
"learning_rate": 4.619377162629758e-05,
|
| 198 |
+
"loss": 2.9617,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2422145328719723,
|
| 203 |
+
"grad_norm": 11.523890495300293,
|
| 204 |
+
"learning_rate": 4.604959630911188e-05,
|
| 205 |
+
"loss": 3.2321,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2508650519031142,
|
| 210 |
+
"grad_norm": 11.910848617553711,
|
| 211 |
+
"learning_rate": 4.5905420991926184e-05,
|
| 212 |
+
"loss": 2.6359,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.25951557093425603,
|
| 217 |
+
"grad_norm": 11.915596008300781,
|
| 218 |
+
"learning_rate": 4.576124567474049e-05,
|
| 219 |
+
"loss": 3.018,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2681660899653979,
|
| 224 |
+
"grad_norm": 25.279979705810547,
|
| 225 |
+
"learning_rate": 4.5617070357554784e-05,
|
| 226 |
+
"loss": 2.9482,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2768166089965398,
|
| 231 |
+
"grad_norm": 19.945743560791016,
|
| 232 |
+
"learning_rate": 4.547289504036909e-05,
|
| 233 |
+
"loss": 3.1401,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.28546712802768165,
|
| 238 |
+
"grad_norm": 15.122772216796875,
|
| 239 |
+
"learning_rate": 4.532871972318339e-05,
|
| 240 |
+
"loss": 2.903,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.29411764705882354,
|
| 245 |
+
"grad_norm": 10.13607406616211,
|
| 246 |
+
"learning_rate": 4.51845444059977e-05,
|
| 247 |
+
"loss": 3.1216,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3027681660899654,
|
| 252 |
+
"grad_norm": 12.250367164611816,
|
| 253 |
+
"learning_rate": 4.504036908881199e-05,
|
| 254 |
+
"loss": 3.1527,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.31141868512110726,
|
| 259 |
+
"grad_norm": 8.994067192077637,
|
| 260 |
+
"learning_rate": 4.48961937716263e-05,
|
| 261 |
+
"loss": 2.931,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.32006920415224915,
|
| 266 |
+
"grad_norm": 9.081120491027832,
|
| 267 |
+
"learning_rate": 4.47520184544406e-05,
|
| 268 |
+
"loss": 2.9034,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.328719723183391,
|
| 273 |
+
"grad_norm": 10.299670219421387,
|
| 274 |
+
"learning_rate": 4.460784313725491e-05,
|
| 275 |
+
"loss": 3.0039,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3373702422145329,
|
| 280 |
+
"grad_norm": 53.3034553527832,
|
| 281 |
+
"learning_rate": 4.446366782006921e-05,
|
| 282 |
+
"loss": 2.9574,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3460207612456747,
|
| 287 |
+
"grad_norm": 10.064446449279785,
|
| 288 |
+
"learning_rate": 4.431949250288351e-05,
|
| 289 |
+
"loss": 2.914,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3546712802768166,
|
| 294 |
+
"grad_norm": 10.004849433898926,
|
| 295 |
+
"learning_rate": 4.4175317185697815e-05,
|
| 296 |
+
"loss": 2.9598,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.3633217993079585,
|
| 301 |
+
"grad_norm": 9.381330490112305,
|
| 302 |
+
"learning_rate": 4.403114186851211e-05,
|
| 303 |
+
"loss": 2.964,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3719723183391003,
|
| 308 |
+
"grad_norm": 6.36283016204834,
|
| 309 |
+
"learning_rate": 4.3886966551326416e-05,
|
| 310 |
+
"loss": 2.8286,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3806228373702422,
|
| 315 |
+
"grad_norm": 9.161887168884277,
|
| 316 |
+
"learning_rate": 4.3742791234140716e-05,
|
| 317 |
+
"loss": 3.1334,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3892733564013841,
|
| 322 |
+
"grad_norm": 7.230731964111328,
|
| 323 |
+
"learning_rate": 4.359861591695502e-05,
|
| 324 |
+
"loss": 3.0322,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.39792387543252594,
|
| 329 |
+
"grad_norm": 17.26264762878418,
|
| 330 |
+
"learning_rate": 4.345444059976932e-05,
|
| 331 |
+
"loss": 3.3673,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.40657439446366783,
|
| 336 |
+
"grad_norm": 11.14217758178711,
|
| 337 |
+
"learning_rate": 4.3310265282583624e-05,
|
| 338 |
+
"loss": 2.9077,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.41522491349480967,
|
| 343 |
+
"grad_norm": 18.250459671020508,
|
| 344 |
+
"learning_rate": 4.3166089965397924e-05,
|
| 345 |
+
"loss": 2.8501,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.42387543252595156,
|
| 350 |
+
"grad_norm": 9.473771095275879,
|
| 351 |
+
"learning_rate": 4.302191464821223e-05,
|
| 352 |
+
"loss": 2.6472,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.43252595155709345,
|
| 357 |
+
"grad_norm": 10.694830894470215,
|
| 358 |
+
"learning_rate": 4.287773933102653e-05,
|
| 359 |
+
"loss": 2.9812,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.4411764705882353,
|
| 364 |
+
"grad_norm": 14.740256309509277,
|
| 365 |
+
"learning_rate": 4.273356401384083e-05,
|
| 366 |
+
"loss": 2.7586,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.44982698961937717,
|
| 371 |
+
"grad_norm": 13.539804458618164,
|
| 372 |
+
"learning_rate": 4.258938869665514e-05,
|
| 373 |
+
"loss": 2.9536,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.458477508650519,
|
| 378 |
+
"grad_norm": 9.57966423034668,
|
| 379 |
+
"learning_rate": 4.244521337946944e-05,
|
| 380 |
+
"loss": 2.5686,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.4671280276816609,
|
| 385 |
+
"grad_norm": 5.894380569458008,
|
| 386 |
+
"learning_rate": 4.230103806228374e-05,
|
| 387 |
+
"loss": 2.9487,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4757785467128028,
|
| 392 |
+
"grad_norm": 6.372859954833984,
|
| 393 |
+
"learning_rate": 4.215686274509804e-05,
|
| 394 |
+
"loss": 3.2582,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4844290657439446,
|
| 399 |
+
"grad_norm": 7.707736015319824,
|
| 400 |
+
"learning_rate": 4.201268742791235e-05,
|
| 401 |
+
"loss": 3.036,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4930795847750865,
|
| 406 |
+
"grad_norm": 12.041340827941895,
|
| 407 |
+
"learning_rate": 4.186851211072664e-05,
|
| 408 |
+
"loss": 3.0721,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5017301038062284,
|
| 413 |
+
"grad_norm": 7.917064666748047,
|
| 414 |
+
"learning_rate": 4.172433679354095e-05,
|
| 415 |
+
"loss": 2.8503,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5103806228373703,
|
| 420 |
+
"grad_norm": 10.359112739562988,
|
| 421 |
+
"learning_rate": 4.158016147635525e-05,
|
| 422 |
+
"loss": 2.7367,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5190311418685121,
|
| 427 |
+
"grad_norm": 27.387939453125,
|
| 428 |
+
"learning_rate": 4.1435986159169555e-05,
|
| 429 |
+
"loss": 2.6254,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.527681660899654,
|
| 434 |
+
"grad_norm": 6.714576244354248,
|
| 435 |
+
"learning_rate": 4.1291810841983856e-05,
|
| 436 |
+
"loss": 2.7978,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5363321799307958,
|
| 441 |
+
"grad_norm": 9.878549575805664,
|
| 442 |
+
"learning_rate": 4.1147635524798156e-05,
|
| 443 |
+
"loss": 2.9624,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5449826989619377,
|
| 448 |
+
"grad_norm": 6.7428131103515625,
|
| 449 |
+
"learning_rate": 4.1003460207612456e-05,
|
| 450 |
+
"loss": 2.9623,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5536332179930796,
|
| 455 |
+
"grad_norm": 13.340039253234863,
|
| 456 |
+
"learning_rate": 4.0859284890426763e-05,
|
| 457 |
+
"loss": 3.0023,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5622837370242214,
|
| 462 |
+
"grad_norm": 9.870621681213379,
|
| 463 |
+
"learning_rate": 4.0715109573241064e-05,
|
| 464 |
+
"loss": 2.8046,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5709342560553633,
|
| 469 |
+
"grad_norm": 6.491202354431152,
|
| 470 |
+
"learning_rate": 4.0570934256055364e-05,
|
| 471 |
+
"loss": 3.1365,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5795847750865052,
|
| 476 |
+
"grad_norm": 12.942968368530273,
|
| 477 |
+
"learning_rate": 4.042675893886967e-05,
|
| 478 |
+
"loss": 2.9377,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5882352941176471,
|
| 483 |
+
"grad_norm": 7.937127590179443,
|
| 484 |
+
"learning_rate": 4.0282583621683965e-05,
|
| 485 |
+
"loss": 2.9007,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.596885813148789,
|
| 490 |
+
"grad_norm": 6.2800421714782715,
|
| 491 |
+
"learning_rate": 4.013840830449827e-05,
|
| 492 |
+
"loss": 2.9504,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6055363321799307,
|
| 497 |
+
"grad_norm": 8.236554145812988,
|
| 498 |
+
"learning_rate": 3.999423298731257e-05,
|
| 499 |
+
"loss": 2.9998,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6141868512110726,
|
| 504 |
+
"grad_norm": 6.5029215812683105,
|
| 505 |
+
"learning_rate": 3.985005767012688e-05,
|
| 506 |
+
"loss": 2.9348,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6228373702422145,
|
| 511 |
+
"grad_norm": 7.156394004821777,
|
| 512 |
+
"learning_rate": 3.970588235294117e-05,
|
| 513 |
+
"loss": 2.5296,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6314878892733564,
|
| 518 |
+
"grad_norm": 6.988163948059082,
|
| 519 |
+
"learning_rate": 3.956170703575548e-05,
|
| 520 |
+
"loss": 2.9235,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6401384083044983,
|
| 525 |
+
"grad_norm": 4.899734020233154,
|
| 526 |
+
"learning_rate": 3.941753171856978e-05,
|
| 527 |
+
"loss": 3.0278,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6487889273356401,
|
| 532 |
+
"grad_norm": 7.569077014923096,
|
| 533 |
+
"learning_rate": 3.927335640138409e-05,
|
| 534 |
+
"loss": 2.8099,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.657439446366782,
|
| 539 |
+
"grad_norm": 10.045110702514648,
|
| 540 |
+
"learning_rate": 3.912918108419839e-05,
|
| 541 |
+
"loss": 2.6677,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6660899653979239,
|
| 546 |
+
"grad_norm": 7.325290679931641,
|
| 547 |
+
"learning_rate": 3.898500576701269e-05,
|
| 548 |
+
"loss": 3.009,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6747404844290658,
|
| 553 |
+
"grad_norm": 4.040036678314209,
|
| 554 |
+
"learning_rate": 3.8840830449826995e-05,
|
| 555 |
+
"loss": 2.8662,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6833910034602076,
|
| 560 |
+
"grad_norm": 24.85966682434082,
|
| 561 |
+
"learning_rate": 3.8696655132641296e-05,
|
| 562 |
+
"loss": 3.0819,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6920415224913494,
|
| 567 |
+
"grad_norm": 7.573343753814697,
|
| 568 |
+
"learning_rate": 3.8552479815455596e-05,
|
| 569 |
+
"loss": 3.1691,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7006920415224913,
|
| 574 |
+
"grad_norm": 5.489699840545654,
|
| 575 |
+
"learning_rate": 3.8408304498269896e-05,
|
| 576 |
+
"loss": 2.7598,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7093425605536332,
|
| 581 |
+
"grad_norm": 6.8325090408325195,
|
| 582 |
+
"learning_rate": 3.8264129181084204e-05,
|
| 583 |
+
"loss": 2.7702,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7179930795847751,
|
| 588 |
+
"grad_norm": 14.738582611083984,
|
| 589 |
+
"learning_rate": 3.81199538638985e-05,
|
| 590 |
+
"loss": 2.5193,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.726643598615917,
|
| 595 |
+
"grad_norm": 5.153214454650879,
|
| 596 |
+
"learning_rate": 3.7975778546712804e-05,
|
| 597 |
+
"loss": 2.6036,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7352941176470589,
|
| 602 |
+
"grad_norm": 5.4641499519348145,
|
| 603 |
+
"learning_rate": 3.7831603229527105e-05,
|
| 604 |
+
"loss": 2.8178,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.7439446366782007,
|
| 609 |
+
"grad_norm": 10.48047924041748,
|
| 610 |
+
"learning_rate": 3.768742791234141e-05,
|
| 611 |
+
"loss": 2.4638,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7525951557093425,
|
| 616 |
+
"grad_norm": 6.817728042602539,
|
| 617 |
+
"learning_rate": 3.754325259515571e-05,
|
| 618 |
+
"loss": 2.7103,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7612456747404844,
|
| 623 |
+
"grad_norm": 11.84115982055664,
|
| 624 |
+
"learning_rate": 3.739907727797001e-05,
|
| 625 |
+
"loss": 2.8477,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7698961937716263,
|
| 630 |
+
"grad_norm": 8.383696556091309,
|
| 631 |
+
"learning_rate": 3.725490196078432e-05,
|
| 632 |
+
"loss": 2.8378,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7785467128027682,
|
| 637 |
+
"grad_norm": 6.298950672149658,
|
| 638 |
+
"learning_rate": 3.711072664359862e-05,
|
| 639 |
+
"loss": 2.968,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.78719723183391,
|
| 644 |
+
"grad_norm": 4.829587936401367,
|
| 645 |
+
"learning_rate": 3.696655132641292e-05,
|
| 646 |
+
"loss": 3.096,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7958477508650519,
|
| 651 |
+
"grad_norm": 4.4047746658325195,
|
| 652 |
+
"learning_rate": 3.682237600922722e-05,
|
| 653 |
+
"loss": 2.7938,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.8044982698961938,
|
| 658 |
+
"grad_norm": 6.090274810791016,
|
| 659 |
+
"learning_rate": 3.667820069204153e-05,
|
| 660 |
+
"loss": 2.6475,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.8131487889273357,
|
| 665 |
+
"grad_norm": 6.233980178833008,
|
| 666 |
+
"learning_rate": 3.653402537485582e-05,
|
| 667 |
+
"loss": 3.033,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.8217993079584776,
|
| 672 |
+
"grad_norm": 6.986254692077637,
|
| 673 |
+
"learning_rate": 3.638985005767013e-05,
|
| 674 |
+
"loss": 2.8964,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.8304498269896193,
|
| 679 |
+
"grad_norm": 6.642579555511475,
|
| 680 |
+
"learning_rate": 3.624567474048443e-05,
|
| 681 |
+
"loss": 2.4588,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8391003460207612,
|
| 686 |
+
"grad_norm": 4.945983409881592,
|
| 687 |
+
"learning_rate": 3.6101499423298736e-05,
|
| 688 |
+
"loss": 2.5413,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8477508650519031,
|
| 693 |
+
"grad_norm": 5.664891719818115,
|
| 694 |
+
"learning_rate": 3.5957324106113036e-05,
|
| 695 |
+
"loss": 2.6916,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.856401384083045,
|
| 700 |
+
"grad_norm": 6.726191520690918,
|
| 701 |
+
"learning_rate": 3.5813148788927336e-05,
|
| 702 |
+
"loss": 2.7688,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8650519031141869,
|
| 707 |
+
"grad_norm": 7.181143760681152,
|
| 708 |
+
"learning_rate": 3.566897347174164e-05,
|
| 709 |
+
"loss": 2.8977,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8737024221453287,
|
| 714 |
+
"grad_norm": 6.649637222290039,
|
| 715 |
+
"learning_rate": 3.5524798154555944e-05,
|
| 716 |
+
"loss": 2.9901,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8823529411764706,
|
| 721 |
+
"grad_norm": 0.0,
|
| 722 |
+
"learning_rate": 3.5380622837370244e-05,
|
| 723 |
+
"loss": 2.6497,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.8910034602076125,
|
| 728 |
+
"grad_norm": 8.609131813049316,
|
| 729 |
+
"learning_rate": 3.5236447520184545e-05,
|
| 730 |
+
"loss": 3.1123,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8996539792387543,
|
| 735 |
+
"grad_norm": 6.280732154846191,
|
| 736 |
+
"learning_rate": 3.509227220299885e-05,
|
| 737 |
+
"loss": 3.0722,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.9083044982698962,
|
| 742 |
+
"grad_norm": 5.108139991760254,
|
| 743 |
+
"learning_rate": 3.494809688581315e-05,
|
| 744 |
+
"loss": 2.8024,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.916955017301038,
|
| 749 |
+
"grad_norm": 6.271091461181641,
|
| 750 |
+
"learning_rate": 3.480392156862745e-05,
|
| 751 |
+
"loss": 3.0201,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.9256055363321799,
|
| 756 |
+
"grad_norm": 6.791084289550781,
|
| 757 |
+
"learning_rate": 3.465974625144175e-05,
|
| 758 |
+
"loss": 2.7122,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.9342560553633218,
|
| 763 |
+
"grad_norm": 14.90443229675293,
|
| 764 |
+
"learning_rate": 3.451557093425606e-05,
|
| 765 |
+
"loss": 2.7771,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.9429065743944637,
|
| 770 |
+
"grad_norm": 7.935113906860352,
|
| 771 |
+
"learning_rate": 3.4371395617070353e-05,
|
| 772 |
+
"loss": 2.8365,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.9515570934256056,
|
| 777 |
+
"grad_norm": 10.18995189666748,
|
| 778 |
+
"learning_rate": 3.422722029988466e-05,
|
| 779 |
+
"loss": 2.8855,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9602076124567474,
|
| 784 |
+
"grad_norm": 7.072559356689453,
|
| 785 |
+
"learning_rate": 3.408304498269896e-05,
|
| 786 |
+
"loss": 2.7104,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9688581314878892,
|
| 791 |
+
"grad_norm": 4.9115400314331055,
|
| 792 |
+
"learning_rate": 3.393886966551327e-05,
|
| 793 |
+
"loss": 2.847,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9775086505190311,
|
| 798 |
+
"grad_norm": 6.05402135848999,
|
| 799 |
+
"learning_rate": 3.379469434832757e-05,
|
| 800 |
+
"loss": 2.4883,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.986159169550173,
|
| 805 |
+
"grad_norm": 6.126327991485596,
|
| 806 |
+
"learning_rate": 3.365051903114187e-05,
|
| 807 |
+
"loss": 2.7995,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9948096885813149,
|
| 812 |
+
"grad_norm": 8.214347839355469,
|
| 813 |
+
"learning_rate": 3.3506343713956176e-05,
|
| 814 |
+
"loss": 2.8806,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.0034602076124568,
|
| 819 |
+
"grad_norm": 6.953145980834961,
|
| 820 |
+
"learning_rate": 3.3362168396770476e-05,
|
| 821 |
+
"loss": 2.6513,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.0121107266435987,
|
| 826 |
+
"grad_norm": 7.279242992401123,
|
| 827 |
+
"learning_rate": 3.3217993079584777e-05,
|
| 828 |
+
"loss": 2.5831,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.0207612456747406,
|
| 833 |
+
"grad_norm": 6.322566032409668,
|
| 834 |
+
"learning_rate": 3.307381776239908e-05,
|
| 835 |
+
"loss": 2.5757,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.0294117647058822,
|
| 840 |
+
"grad_norm": 5.4494547843933105,
|
| 841 |
+
"learning_rate": 3.2929642445213384e-05,
|
| 842 |
+
"loss": 2.3202,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.0380622837370241,
|
| 847 |
+
"grad_norm": 0.0,
|
| 848 |
+
"learning_rate": 3.2785467128027684e-05,
|
| 849 |
+
"loss": 2.1757,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.046712802768166,
|
| 854 |
+
"grad_norm": 6.834903717041016,
|
| 855 |
+
"learning_rate": 3.2641291810841985e-05,
|
| 856 |
+
"loss": 2.4916,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.055363321799308,
|
| 861 |
+
"grad_norm": 7.15425443649292,
|
| 862 |
+
"learning_rate": 3.2497116493656285e-05,
|
| 863 |
+
"loss": 2.5786,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.0640138408304498,
|
| 868 |
+
"grad_norm": 25.43375587463379,
|
| 869 |
+
"learning_rate": 3.235294117647059e-05,
|
| 870 |
+
"loss": 2.457,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.0726643598615917,
|
| 875 |
+
"grad_norm": 6.561356067657471,
|
| 876 |
+
"learning_rate": 3.220876585928489e-05,
|
| 877 |
+
"loss": 2.3607,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.0813148788927336,
|
| 882 |
+
"grad_norm": 11.72268009185791,
|
| 883 |
+
"learning_rate": 3.206459054209919e-05,
|
| 884 |
+
"loss": 2.3018,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.0899653979238755,
|
| 889 |
+
"grad_norm": 0.0,
|
| 890 |
+
"learning_rate": 3.19204152249135e-05,
|
| 891 |
+
"loss": 2.4857,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.0986159169550174,
|
| 896 |
+
"grad_norm": 5.507035255432129,
|
| 897 |
+
"learning_rate": 3.17762399077278e-05,
|
| 898 |
+
"loss": 2.4546,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.1072664359861593,
|
| 903 |
+
"grad_norm": 8.205038070678711,
|
| 904 |
+
"learning_rate": 3.16320645905421e-05,
|
| 905 |
+
"loss": 2.5896,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.1159169550173011,
|
| 910 |
+
"grad_norm": 6.336702346801758,
|
| 911 |
+
"learning_rate": 3.14878892733564e-05,
|
| 912 |
+
"loss": 2.6643,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.1245674740484428,
|
| 917 |
+
"grad_norm": 6.227362155914307,
|
| 918 |
+
"learning_rate": 3.134371395617071e-05,
|
| 919 |
+
"loss": 2.4297,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.1332179930795847,
|
| 924 |
+
"grad_norm": 7.871458053588867,
|
| 925 |
+
"learning_rate": 3.119953863898501e-05,
|
| 926 |
+
"loss": 2.5482,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.1418685121107266,
|
| 931 |
+
"grad_norm": 5.489798545837402,
|
| 932 |
+
"learning_rate": 3.105536332179931e-05,
|
| 933 |
+
"loss": 2.3449,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.1505190311418685,
|
| 938 |
+
"grad_norm": 4.873505592346191,
|
| 939 |
+
"learning_rate": 3.091118800461361e-05,
|
| 940 |
+
"loss": 2.3388,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.1591695501730104,
|
| 945 |
+
"grad_norm": 7.125423431396484,
|
| 946 |
+
"learning_rate": 3.0767012687427916e-05,
|
| 947 |
+
"loss": 2.3973,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.1678200692041523,
|
| 952 |
+
"grad_norm": 8.925414085388184,
|
| 953 |
+
"learning_rate": 3.062283737024222e-05,
|
| 954 |
+
"loss": 2.3922,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.1764705882352942,
|
| 959 |
+
"grad_norm": 11.737508773803711,
|
| 960 |
+
"learning_rate": 3.0478662053056517e-05,
|
| 961 |
+
"loss": 2.4235,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.185121107266436,
|
| 966 |
+
"grad_norm": 7.375851154327393,
|
| 967 |
+
"learning_rate": 3.033448673587082e-05,
|
| 968 |
+
"loss": 2.3064,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.193771626297578,
|
| 973 |
+
"grad_norm": 9.243279457092285,
|
| 974 |
+
"learning_rate": 3.0190311418685124e-05,
|
| 975 |
+
"loss": 2.2488,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.2024221453287196,
|
| 980 |
+
"grad_norm": 8.473196983337402,
|
| 981 |
+
"learning_rate": 3.004613610149942e-05,
|
| 982 |
+
"loss": 2.4362,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.2110726643598615,
|
| 987 |
+
"grad_norm": 4.9002790451049805,
|
| 988 |
+
"learning_rate": 2.9901960784313725e-05,
|
| 989 |
+
"loss": 2.5061,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.2197231833910034,
|
| 994 |
+
"grad_norm": 8.8832368850708,
|
| 995 |
+
"learning_rate": 2.975778546712803e-05,
|
| 996 |
+
"loss": 2.3609,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.2283737024221453,
|
| 1001 |
+
"grad_norm": 9.568990707397461,
|
| 1002 |
+
"learning_rate": 2.9613610149942333e-05,
|
| 1003 |
+
"loss": 2.4853,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.2370242214532872,
|
| 1008 |
+
"grad_norm": 10.570289611816406,
|
| 1009 |
+
"learning_rate": 2.9469434832756633e-05,
|
| 1010 |
+
"loss": 2.3944,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.245674740484429,
|
| 1015 |
+
"grad_norm": 10.139781951904297,
|
| 1016 |
+
"learning_rate": 2.9325259515570937e-05,
|
| 1017 |
+
"loss": 2.4187,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.254325259515571,
|
| 1022 |
+
"grad_norm": 12.260625839233398,
|
| 1023 |
+
"learning_rate": 2.918108419838524e-05,
|
| 1024 |
+
"loss": 2.3519,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.2629757785467128,
|
| 1029 |
+
"grad_norm": 7.253377437591553,
|
| 1030 |
+
"learning_rate": 2.9036908881199544e-05,
|
| 1031 |
+
"loss": 2.4657,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.2716262975778547,
|
| 1036 |
+
"grad_norm": 4.9923553466796875,
|
| 1037 |
+
"learning_rate": 2.889273356401384e-05,
|
| 1038 |
+
"loss": 2.4163,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.2802768166089966,
|
| 1043 |
+
"grad_norm": 9.419482231140137,
|
| 1044 |
+
"learning_rate": 2.8748558246828145e-05,
|
| 1045 |
+
"loss": 2.3895,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.2889273356401385,
|
| 1050 |
+
"grad_norm": 5.093005180358887,
|
| 1051 |
+
"learning_rate": 2.860438292964245e-05,
|
| 1052 |
+
"loss": 2.3857,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.2975778546712804,
|
| 1057 |
+
"grad_norm": 14.546469688415527,
|
| 1058 |
+
"learning_rate": 2.8460207612456746e-05,
|
| 1059 |
+
"loss": 2.3892,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.306228373702422,
|
| 1064 |
+
"grad_norm": 4.272151470184326,
|
| 1065 |
+
"learning_rate": 2.831603229527105e-05,
|
| 1066 |
+
"loss": 2.7683,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.314878892733564,
|
| 1071 |
+
"grad_norm": 6.617842197418213,
|
| 1072 |
+
"learning_rate": 2.8171856978085353e-05,
|
| 1073 |
+
"loss": 2.3034,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.3235294117647058,
|
| 1078 |
+
"grad_norm": 5.439128398895264,
|
| 1079 |
+
"learning_rate": 2.8027681660899657e-05,
|
| 1080 |
+
"loss": 2.6287,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.3321799307958477,
|
| 1085 |
+
"grad_norm": 6.48740816116333,
|
| 1086 |
+
"learning_rate": 2.7883506343713957e-05,
|
| 1087 |
+
"loss": 2.5707,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.3408304498269896,
|
| 1092 |
+
"grad_norm": 6.10065221786499,
|
| 1093 |
+
"learning_rate": 2.773933102652826e-05,
|
| 1094 |
+
"loss": 2.3402,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.3494809688581315,
|
| 1099 |
+
"grad_norm": 5.562924861907959,
|
| 1100 |
+
"learning_rate": 2.759515570934256e-05,
|
| 1101 |
+
"loss": 2.3861,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.3581314878892734,
|
| 1106 |
+
"grad_norm": 8.798916816711426,
|
| 1107 |
+
"learning_rate": 2.7450980392156865e-05,
|
| 1108 |
+
"loss": 2.3875,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.3667820069204153,
|
| 1113 |
+
"grad_norm": 13.19672679901123,
|
| 1114 |
+
"learning_rate": 2.7306805074971165e-05,
|
| 1115 |
+
"loss": 2.1105,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.375432525951557,
|
| 1120 |
+
"grad_norm": 7.34409236907959,
|
| 1121 |
+
"learning_rate": 2.716262975778547e-05,
|
| 1122 |
+
"loss": 2.3937,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.3840830449826989,
|
| 1127 |
+
"grad_norm": 6.497256278991699,
|
| 1128 |
+
"learning_rate": 2.7018454440599773e-05,
|
| 1129 |
+
"loss": 2.3291,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.3927335640138407,
|
| 1134 |
+
"grad_norm": 4.751974582672119,
|
| 1135 |
+
"learning_rate": 2.687427912341407e-05,
|
| 1136 |
+
"loss": 2.528,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.4013840830449826,
|
| 1141 |
+
"grad_norm": 12.393454551696777,
|
| 1142 |
+
"learning_rate": 2.6730103806228373e-05,
|
| 1143 |
+
"loss": 2.5473,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.4100346020761245,
|
| 1148 |
+
"grad_norm": 5.408919811248779,
|
| 1149 |
+
"learning_rate": 2.6585928489042677e-05,
|
| 1150 |
+
"loss": 2.3739,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.4186851211072664,
|
| 1155 |
+
"grad_norm": 6.614011287689209,
|
| 1156 |
+
"learning_rate": 2.644175317185698e-05,
|
| 1157 |
+
"loss": 2.3951,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.4273356401384083,
|
| 1162 |
+
"grad_norm": 6.63184928894043,
|
| 1163 |
+
"learning_rate": 2.6297577854671278e-05,
|
| 1164 |
+
"loss": 2.3318,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.4359861591695502,
|
| 1169 |
+
"grad_norm": 6.916550636291504,
|
| 1170 |
+
"learning_rate": 2.615340253748558e-05,
|
| 1171 |
+
"loss": 2.5457,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.444636678200692,
|
| 1176 |
+
"grad_norm": 7.225638389587402,
|
| 1177 |
+
"learning_rate": 2.6009227220299885e-05,
|
| 1178 |
+
"loss": 2.4932,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.453287197231834,
|
| 1183 |
+
"grad_norm": 7.826704978942871,
|
| 1184 |
+
"learning_rate": 2.586505190311419e-05,
|
| 1185 |
+
"loss": 2.3264,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.4619377162629759,
|
| 1190 |
+
"grad_norm": 12.383480072021484,
|
| 1191 |
+
"learning_rate": 2.572087658592849e-05,
|
| 1192 |
+
"loss": 2.3894,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.4705882352941178,
|
| 1197 |
+
"grad_norm": 6.668808937072754,
|
| 1198 |
+
"learning_rate": 2.5576701268742793e-05,
|
| 1199 |
+
"loss": 2.4778,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.4792387543252596,
|
| 1204 |
+
"grad_norm": 7.074410915374756,
|
| 1205 |
+
"learning_rate": 2.5432525951557097e-05,
|
| 1206 |
+
"loss": 2.166,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.4878892733564013,
|
| 1211 |
+
"grad_norm": 9.129437446594238,
|
| 1212 |
+
"learning_rate": 2.52883506343714e-05,
|
| 1213 |
+
"loss": 2.2821,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.4965397923875432,
|
| 1218 |
+
"grad_norm": 7.847931385040283,
|
| 1219 |
+
"learning_rate": 2.5144175317185697e-05,
|
| 1220 |
+
"loss": 2.2413,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.505190311418685,
|
| 1225 |
+
"grad_norm": 6.493180274963379,
|
| 1226 |
+
"learning_rate": 2.5e-05,
|
| 1227 |
+
"loss": 2.3985,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.513840830449827,
|
| 1232 |
+
"grad_norm": 8.56859016418457,
|
| 1233 |
+
"learning_rate": 2.48558246828143e-05,
|
| 1234 |
+
"loss": 2.4284,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.5224913494809689,
|
| 1239 |
+
"grad_norm": 0.0,
|
| 1240 |
+
"learning_rate": 2.4711649365628605e-05,
|
| 1241 |
+
"loss": 2.5876,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.5311418685121108,
|
| 1246 |
+
"grad_norm": 7.028433799743652,
|
| 1247 |
+
"learning_rate": 2.4567474048442906e-05,
|
| 1248 |
+
"loss": 2.4459,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.5397923875432526,
|
| 1253 |
+
"grad_norm": 5.476517200469971,
|
| 1254 |
+
"learning_rate": 2.442329873125721e-05,
|
| 1255 |
+
"loss": 2.3663,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.5484429065743943,
|
| 1260 |
+
"grad_norm": 7.162520408630371,
|
| 1261 |
+
"learning_rate": 2.427912341407151e-05,
|
| 1262 |
+
"loss": 2.2765,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.5570934256055362,
|
| 1267 |
+
"grad_norm": 13.420760154724121,
|
| 1268 |
+
"learning_rate": 2.4134948096885813e-05,
|
| 1269 |
+
"loss": 2.2691,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.565743944636678,
|
| 1274 |
+
"grad_norm": 7.58514928817749,
|
| 1275 |
+
"learning_rate": 2.3990772779700117e-05,
|
| 1276 |
+
"loss": 2.4497,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.57439446366782,
|
| 1281 |
+
"grad_norm": 5.244626998901367,
|
| 1282 |
+
"learning_rate": 2.384659746251442e-05,
|
| 1283 |
+
"loss": 2.2665,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.5830449826989619,
|
| 1288 |
+
"grad_norm": 8.593738555908203,
|
| 1289 |
+
"learning_rate": 2.370242214532872e-05,
|
| 1290 |
+
"loss": 2.5301,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.5916955017301038,
|
| 1295 |
+
"grad_norm": 5.1958231925964355,
|
| 1296 |
+
"learning_rate": 2.3558246828143025e-05,
|
| 1297 |
+
"loss": 2.2558,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.6003460207612457,
|
| 1302 |
+
"grad_norm": 5.891276836395264,
|
| 1303 |
+
"learning_rate": 2.3414071510957325e-05,
|
| 1304 |
+
"loss": 2.437,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.6089965397923875,
|
| 1309 |
+
"grad_norm": 8.773421287536621,
|
| 1310 |
+
"learning_rate": 2.326989619377163e-05,
|
| 1311 |
+
"loss": 2.2866,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.6176470588235294,
|
| 1316 |
+
"grad_norm": 5.609074592590332,
|
| 1317 |
+
"learning_rate": 2.312572087658593e-05,
|
| 1318 |
+
"loss": 2.3159,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.6262975778546713,
|
| 1323 |
+
"grad_norm": 7.042026519775391,
|
| 1324 |
+
"learning_rate": 2.298154555940023e-05,
|
| 1325 |
+
"loss": 2.4329,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.6349480968858132,
|
| 1330 |
+
"grad_norm": 5.367044925689697,
|
| 1331 |
+
"learning_rate": 2.2837370242214533e-05,
|
| 1332 |
+
"loss": 2.2781,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.643598615916955,
|
| 1337 |
+
"grad_norm": 5.378485202789307,
|
| 1338 |
+
"learning_rate": 2.2693194925028834e-05,
|
| 1339 |
+
"loss": 2.3279,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.652249134948097,
|
| 1344 |
+
"grad_norm": 8.597953796386719,
|
| 1345 |
+
"learning_rate": 2.2549019607843138e-05,
|
| 1346 |
+
"loss": 2.5955,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.6608996539792389,
|
| 1351 |
+
"grad_norm": 7.4986572265625,
|
| 1352 |
+
"learning_rate": 2.240484429065744e-05,
|
| 1353 |
+
"loss": 2.25,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.6695501730103808,
|
| 1358 |
+
"grad_norm": 12.225547790527344,
|
| 1359 |
+
"learning_rate": 2.226066897347174e-05,
|
| 1360 |
+
"loss": 2.1325,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.6782006920415224,
|
| 1365 |
+
"grad_norm": 25.887487411499023,
|
| 1366 |
+
"learning_rate": 2.2116493656286045e-05,
|
| 1367 |
+
"loss": 2.3929,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.6868512110726643,
|
| 1372 |
+
"grad_norm": 13.56473445892334,
|
| 1373 |
+
"learning_rate": 2.197231833910035e-05,
|
| 1374 |
+
"loss": 2.3572,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.6955017301038062,
|
| 1379 |
+
"grad_norm": 10.653745651245117,
|
| 1380 |
+
"learning_rate": 2.182814302191465e-05,
|
| 1381 |
+
"loss": 2.4658,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.7041522491349481,
|
| 1386 |
+
"grad_norm": 9.479430198669434,
|
| 1387 |
+
"learning_rate": 2.1683967704728953e-05,
|
| 1388 |
+
"loss": 2.4963,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.71280276816609,
|
| 1393 |
+
"grad_norm": 12.149545669555664,
|
| 1394 |
+
"learning_rate": 2.1539792387543253e-05,
|
| 1395 |
+
"loss": 2.2464,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.7214532871972317,
|
| 1400 |
+
"grad_norm": 8.545853614807129,
|
| 1401 |
+
"learning_rate": 2.1395617070357557e-05,
|
| 1402 |
+
"loss": 2.3789,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.7301038062283736,
|
| 1407 |
+
"grad_norm": 10.555283546447754,
|
| 1408 |
+
"learning_rate": 2.1251441753171858e-05,
|
| 1409 |
+
"loss": 2.2933,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.7387543252595155,
|
| 1414 |
+
"grad_norm": 10.031829833984375,
|
| 1415 |
+
"learning_rate": 2.1107266435986158e-05,
|
| 1416 |
+
"loss": 2.3333,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.7474048442906573,
|
| 1421 |
+
"grad_norm": 0.0,
|
| 1422 |
+
"learning_rate": 2.096309111880046e-05,
|
| 1423 |
+
"loss": 2.7103,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.7560553633217992,
|
| 1428 |
+
"grad_norm": 7.1533522605896,
|
| 1429 |
+
"learning_rate": 2.0818915801614762e-05,
|
| 1430 |
+
"loss": 2.1076,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.7647058823529411,
|
| 1435 |
+
"grad_norm": 11.710586547851562,
|
| 1436 |
+
"learning_rate": 2.0674740484429066e-05,
|
| 1437 |
+
"loss": 2.352,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.773356401384083,
|
| 1442 |
+
"grad_norm": 10.389070510864258,
|
| 1443 |
+
"learning_rate": 2.053056516724337e-05,
|
| 1444 |
+
"loss": 2.5781,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.782006920415225,
|
| 1449 |
+
"grad_norm": 10.627307891845703,
|
| 1450 |
+
"learning_rate": 2.0386389850057673e-05,
|
| 1451 |
+
"loss": 2.4335,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.7906574394463668,
|
| 1456 |
+
"grad_norm": 7.498691082000732,
|
| 1457 |
+
"learning_rate": 2.0242214532871974e-05,
|
| 1458 |
+
"loss": 2.4912,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.7993079584775087,
|
| 1463 |
+
"grad_norm": 7.602413654327393,
|
| 1464 |
+
"learning_rate": 2.0098039215686277e-05,
|
| 1465 |
+
"loss": 2.396,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.8079584775086506,
|
| 1470 |
+
"grad_norm": 5.651869297027588,
|
| 1471 |
+
"learning_rate": 1.9953863898500578e-05,
|
| 1472 |
+
"loss": 2.2124,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.8166089965397925,
|
| 1477 |
+
"grad_norm": 5.149964809417725,
|
| 1478 |
+
"learning_rate": 1.980968858131488e-05,
|
| 1479 |
+
"loss": 2.47,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.8252595155709344,
|
| 1484 |
+
"grad_norm": 5.61548376083374,
|
| 1485 |
+
"learning_rate": 1.966551326412918e-05,
|
| 1486 |
+
"loss": 2.3634,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.8339100346020762,
|
| 1491 |
+
"grad_norm": 6.321987628936768,
|
| 1492 |
+
"learning_rate": 1.9521337946943485e-05,
|
| 1493 |
+
"loss": 2.3837,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.8425605536332181,
|
| 1498 |
+
"grad_norm": 17.681230545043945,
|
| 1499 |
+
"learning_rate": 1.9377162629757786e-05,
|
| 1500 |
+
"loss": 2.5887,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.85121107266436,
|
| 1505 |
+
"grad_norm": 9.560349464416504,
|
| 1506 |
+
"learning_rate": 1.923298731257209e-05,
|
| 1507 |
+
"loss": 2.6304,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.8598615916955017,
|
| 1512 |
+
"grad_norm": 4.678035736083984,
|
| 1513 |
+
"learning_rate": 1.908881199538639e-05,
|
| 1514 |
+
"loss": 2.3547,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.8685121107266436,
|
| 1519 |
+
"grad_norm": 6.619549751281738,
|
| 1520 |
+
"learning_rate": 1.894463667820069e-05,
|
| 1521 |
+
"loss": 2.4747,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.8771626297577855,
|
| 1526 |
+
"grad_norm": 12.084946632385254,
|
| 1527 |
+
"learning_rate": 1.8800461361014994e-05,
|
| 1528 |
+
"loss": 2.46,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.8858131487889274,
|
| 1533 |
+
"grad_norm": 8.923602104187012,
|
| 1534 |
+
"learning_rate": 1.8656286043829298e-05,
|
| 1535 |
+
"loss": 2.3937,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.8944636678200693,
|
| 1540 |
+
"grad_norm": 9.561132431030273,
|
| 1541 |
+
"learning_rate": 1.85121107266436e-05,
|
| 1542 |
+
"loss": 2.1683,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.903114186851211,
|
| 1547 |
+
"grad_norm": 6.697421073913574,
|
| 1548 |
+
"learning_rate": 1.8367935409457902e-05,
|
| 1549 |
+
"loss": 2.3719,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.9117647058823528,
|
| 1554 |
+
"grad_norm": 6.0673322677612305,
|
| 1555 |
+
"learning_rate": 1.8223760092272205e-05,
|
| 1556 |
+
"loss": 2.1131,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.9204152249134947,
|
| 1561 |
+
"grad_norm": 6.221293926239014,
|
| 1562 |
+
"learning_rate": 1.8079584775086506e-05,
|
| 1563 |
+
"loss": 2.3373,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.9290657439446366,
|
| 1568 |
+
"grad_norm": 19.771453857421875,
|
| 1569 |
+
"learning_rate": 1.793540945790081e-05,
|
| 1570 |
+
"loss": 2.1739,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.9377162629757785,
|
| 1575 |
+
"grad_norm": 4.692147731781006,
|
| 1576 |
+
"learning_rate": 1.779123414071511e-05,
|
| 1577 |
+
"loss": 2.3212,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.9463667820069204,
|
| 1582 |
+
"grad_norm": 13.161782264709473,
|
| 1583 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 1584 |
+
"loss": 2.3698,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.9550173010380623,
|
| 1589 |
+
"grad_norm": 7.448920726776123,
|
| 1590 |
+
"learning_rate": 1.7502883506343714e-05,
|
| 1591 |
+
"loss": 2.0373,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.9636678200692042,
|
| 1596 |
+
"grad_norm": 8.207565307617188,
|
| 1597 |
+
"learning_rate": 1.7358708189158018e-05,
|
| 1598 |
+
"loss": 2.4333,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.972318339100346,
|
| 1603 |
+
"grad_norm": 7.432470798492432,
|
| 1604 |
+
"learning_rate": 1.7214532871972318e-05,
|
| 1605 |
+
"loss": 2.1247,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.980968858131488,
|
| 1610 |
+
"grad_norm": 8.950566291809082,
|
| 1611 |
+
"learning_rate": 1.7070357554786622e-05,
|
| 1612 |
+
"loss": 2.2152,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.9896193771626298,
|
| 1617 |
+
"grad_norm": 8.370878219604492,
|
| 1618 |
+
"learning_rate": 1.6926182237600925e-05,
|
| 1619 |
+
"loss": 2.3552,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.9982698961937717,
|
| 1624 |
+
"grad_norm": 9.798127174377441,
|
| 1625 |
+
"learning_rate": 1.6782006920415226e-05,
|
| 1626 |
+
"loss": 2.5178,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 2.0069204152249136,
|
| 1631 |
+
"grad_norm": 4.229007244110107,
|
| 1632 |
+
"learning_rate": 1.663783160322953e-05,
|
| 1633 |
+
"loss": 2.1752,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.0155709342560555,
|
| 1638 |
+
"grad_norm": 6.5957255363464355,
|
| 1639 |
+
"learning_rate": 1.649365628604383e-05,
|
| 1640 |
+
"loss": 2.2877,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 2.0242214532871974,
|
| 1645 |
+
"grad_norm": 6.140663146972656,
|
| 1646 |
+
"learning_rate": 1.6349480968858134e-05,
|
| 1647 |
+
"loss": 2.2057,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.0328719723183393,
|
| 1652 |
+
"grad_norm": 11.750750541687012,
|
| 1653 |
+
"learning_rate": 1.6205305651672434e-05,
|
| 1654 |
+
"loss": 2.1969,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 2.041522491349481,
|
| 1659 |
+
"grad_norm": 8.632787704467773,
|
| 1660 |
+
"learning_rate": 1.6061130334486738e-05,
|
| 1661 |
+
"loss": 2.1248,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 2.050173010380623,
|
| 1666 |
+
"grad_norm": 13.063236236572266,
|
| 1667 |
+
"learning_rate": 1.5916955017301038e-05,
|
| 1668 |
+
"loss": 2.056,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 2.0588235294117645,
|
| 1673 |
+
"grad_norm": 16.210952758789062,
|
| 1674 |
+
"learning_rate": 1.5772779700115342e-05,
|
| 1675 |
+
"loss": 1.8912,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.0674740484429064,
|
| 1680 |
+
"grad_norm": 13.75781536102295,
|
| 1681 |
+
"learning_rate": 1.5628604382929642e-05,
|
| 1682 |
+
"loss": 2.0829,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.0761245674740483,
|
| 1687 |
+
"grad_norm": 6.415799617767334,
|
| 1688 |
+
"learning_rate": 1.5484429065743946e-05,
|
| 1689 |
+
"loss": 2.0569,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.08477508650519,
|
| 1694 |
+
"grad_norm": 7.936450004577637,
|
| 1695 |
+
"learning_rate": 1.5340253748558246e-05,
|
| 1696 |
+
"loss": 2.1501,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.093425605536332,
|
| 1701 |
+
"grad_norm": 7.677123069763184,
|
| 1702 |
+
"learning_rate": 1.5196078431372548e-05,
|
| 1703 |
+
"loss": 1.9873,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.102076124567474,
|
| 1708 |
+
"grad_norm": 7.399821758270264,
|
| 1709 |
+
"learning_rate": 1.5051903114186852e-05,
|
| 1710 |
+
"loss": 2.078,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.110726643598616,
|
| 1715 |
+
"grad_norm": 9.003458976745605,
|
| 1716 |
+
"learning_rate": 1.4907727797001152e-05,
|
| 1717 |
+
"loss": 1.9666,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.1193771626297577,
|
| 1722 |
+
"grad_norm": 5.406769752502441,
|
| 1723 |
+
"learning_rate": 1.4763552479815456e-05,
|
| 1724 |
+
"loss": 2.1621,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.1280276816608996,
|
| 1729 |
+
"grad_norm": 9.71753978729248,
|
| 1730 |
+
"learning_rate": 1.4619377162629758e-05,
|
| 1731 |
+
"loss": 2.3213,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.1366782006920415,
|
| 1736 |
+
"grad_norm": 7.547745227813721,
|
| 1737 |
+
"learning_rate": 1.4475201845444062e-05,
|
| 1738 |
+
"loss": 1.981,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.1453287197231834,
|
| 1743 |
+
"grad_norm": 6.105127811431885,
|
| 1744 |
+
"learning_rate": 1.4331026528258362e-05,
|
| 1745 |
+
"loss": 2.0367,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.1539792387543253,
|
| 1750 |
+
"grad_norm": 19.412294387817383,
|
| 1751 |
+
"learning_rate": 1.4186851211072666e-05,
|
| 1752 |
+
"loss": 2.0869,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.162629757785467,
|
| 1757 |
+
"grad_norm": 7.70440149307251,
|
| 1758 |
+
"learning_rate": 1.4042675893886966e-05,
|
| 1759 |
+
"loss": 2.1693,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.171280276816609,
|
| 1764 |
+
"grad_norm": 7.652112007141113,
|
| 1765 |
+
"learning_rate": 1.389850057670127e-05,
|
| 1766 |
+
"loss": 2.1045,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.179930795847751,
|
| 1771 |
+
"grad_norm": 11.779850959777832,
|
| 1772 |
+
"learning_rate": 1.3754325259515572e-05,
|
| 1773 |
+
"loss": 2.0855,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.188581314878893,
|
| 1778 |
+
"grad_norm": 0.0,
|
| 1779 |
+
"learning_rate": 1.3610149942329876e-05,
|
| 1780 |
+
"loss": 2.1312,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.1972318339100347,
|
| 1785 |
+
"grad_norm": 6.408977031707764,
|
| 1786 |
+
"learning_rate": 1.3465974625144176e-05,
|
| 1787 |
+
"loss": 2.1698,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.2058823529411766,
|
| 1792 |
+
"grad_norm": 8.69074821472168,
|
| 1793 |
+
"learning_rate": 1.3321799307958476e-05,
|
| 1794 |
+
"loss": 2.0696,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.2145328719723185,
|
| 1799 |
+
"grad_norm": 8.960659980773926,
|
| 1800 |
+
"learning_rate": 1.317762399077278e-05,
|
| 1801 |
+
"loss": 1.73,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.2231833910034604,
|
| 1806 |
+
"grad_norm": 8.215168952941895,
|
| 1807 |
+
"learning_rate": 1.3033448673587082e-05,
|
| 1808 |
+
"loss": 2.0223,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.2318339100346023,
|
| 1813 |
+
"grad_norm": 6.832982063293457,
|
| 1814 |
+
"learning_rate": 1.2889273356401386e-05,
|
| 1815 |
+
"loss": 2.2219,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.2404844290657437,
|
| 1820 |
+
"grad_norm": 15.58371639251709,
|
| 1821 |
+
"learning_rate": 1.2745098039215686e-05,
|
| 1822 |
+
"loss": 2.1099,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.2491349480968856,
|
| 1827 |
+
"grad_norm": 5.2607197761535645,
|
| 1828 |
+
"learning_rate": 1.260092272202999e-05,
|
| 1829 |
+
"loss": 2.1367,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.2577854671280275,
|
| 1834 |
+
"grad_norm": 5.739188194274902,
|
| 1835 |
+
"learning_rate": 1.245674740484429e-05,
|
| 1836 |
+
"loss": 2.2481,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.2664359861591694,
|
| 1841 |
+
"grad_norm": 0.0,
|
| 1842 |
+
"learning_rate": 1.2312572087658592e-05,
|
| 1843 |
+
"loss": 2.1261,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.2750865051903113,
|
| 1848 |
+
"grad_norm": 15.587763786315918,
|
| 1849 |
+
"learning_rate": 1.2168396770472894e-05,
|
| 1850 |
+
"loss": 2.301,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.283737024221453,
|
| 1855 |
+
"grad_norm": 4.138315200805664,
|
| 1856 |
+
"learning_rate": 1.2024221453287198e-05,
|
| 1857 |
+
"loss": 2.2832,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.292387543252595,
|
| 1862 |
+
"grad_norm": 7.0834784507751465,
|
| 1863 |
+
"learning_rate": 1.18800461361015e-05,
|
| 1864 |
+
"loss": 2.2382,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.301038062283737,
|
| 1869 |
+
"grad_norm": 7.361220359802246,
|
| 1870 |
+
"learning_rate": 1.1735870818915802e-05,
|
| 1871 |
+
"loss": 2.0817,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.309688581314879,
|
| 1876 |
+
"grad_norm": 15.36158275604248,
|
| 1877 |
+
"learning_rate": 1.1591695501730104e-05,
|
| 1878 |
+
"loss": 1.8927,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.3183391003460208,
|
| 1883 |
+
"grad_norm": 5.309439659118652,
|
| 1884 |
+
"learning_rate": 1.1447520184544406e-05,
|
| 1885 |
+
"loss": 1.9435,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.3269896193771626,
|
| 1890 |
+
"grad_norm": 0.0,
|
| 1891 |
+
"learning_rate": 1.1303344867358708e-05,
|
| 1892 |
+
"loss": 1.9827,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.3356401384083045,
|
| 1897 |
+
"grad_norm": 5.341144561767578,
|
| 1898 |
+
"learning_rate": 1.115916955017301e-05,
|
| 1899 |
+
"loss": 2.2153,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.3442906574394464,
|
| 1904 |
+
"grad_norm": 7.391653537750244,
|
| 1905 |
+
"learning_rate": 1.1014994232987314e-05,
|
| 1906 |
+
"loss": 2.0618,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.3529411764705883,
|
| 1911 |
+
"grad_norm": 11.139851570129395,
|
| 1912 |
+
"learning_rate": 1.0870818915801616e-05,
|
| 1913 |
+
"loss": 2.02,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.36159169550173,
|
| 1918 |
+
"grad_norm": 6.582757472991943,
|
| 1919 |
+
"learning_rate": 1.0726643598615918e-05,
|
| 1920 |
+
"loss": 2.2848,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.370242214532872,
|
| 1925 |
+
"grad_norm": 8.014312744140625,
|
| 1926 |
+
"learning_rate": 1.058246828143022e-05,
|
| 1927 |
+
"loss": 2.3369,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.378892733564014,
|
| 1932 |
+
"grad_norm": 17.67845344543457,
|
| 1933 |
+
"learning_rate": 1.043829296424452e-05,
|
| 1934 |
+
"loss": 2.1467,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.387543252595156,
|
| 1939 |
+
"grad_norm": 4.640291690826416,
|
| 1940 |
+
"learning_rate": 1.0294117647058824e-05,
|
| 1941 |
+
"loss": 2.2248,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.3961937716262978,
|
| 1946 |
+
"grad_norm": 10.063308715820312,
|
| 1947 |
+
"learning_rate": 1.0149942329873126e-05,
|
| 1948 |
+
"loss": 2.1502,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.404844290657439,
|
| 1953 |
+
"grad_norm": 10.428187370300293,
|
| 1954 |
+
"learning_rate": 1.0005767012687428e-05,
|
| 1955 |
+
"loss": 2.1736,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.413494809688581,
|
| 1960 |
+
"grad_norm": 5.396498680114746,
|
| 1961 |
+
"learning_rate": 9.86159169550173e-06,
|
| 1962 |
+
"loss": 1.9383,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.422145328719723,
|
| 1967 |
+
"grad_norm": 0.0,
|
| 1968 |
+
"learning_rate": 9.717416378316032e-06,
|
| 1969 |
+
"loss": 1.9406,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.430795847750865,
|
| 1974 |
+
"grad_norm": 8.664689064025879,
|
| 1975 |
+
"learning_rate": 9.573241061130335e-06,
|
| 1976 |
+
"loss": 2.3603,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.4394463667820068,
|
| 1981 |
+
"grad_norm": 13.918140411376953,
|
| 1982 |
+
"learning_rate": 9.429065743944637e-06,
|
| 1983 |
+
"loss": 2.2744,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.4480968858131487,
|
| 1988 |
+
"grad_norm": 0.0,
|
| 1989 |
+
"learning_rate": 9.28489042675894e-06,
|
| 1990 |
+
"loss": 2.1881,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.4567474048442905,
|
| 1995 |
+
"grad_norm": 5.737022876739502,
|
| 1996 |
+
"learning_rate": 9.140715109573242e-06,
|
| 1997 |
+
"loss": 2.0455,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.4653979238754324,
|
| 2002 |
+
"grad_norm": 6.97428560256958,
|
| 2003 |
+
"learning_rate": 8.996539792387544e-06,
|
| 2004 |
+
"loss": 1.9571,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.4740484429065743,
|
| 2009 |
+
"grad_norm": 5.972537994384766,
|
| 2010 |
+
"learning_rate": 8.852364475201846e-06,
|
| 2011 |
+
"loss": 2.2661,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.482698961937716,
|
| 2016 |
+
"grad_norm": 8.718277931213379,
|
| 2017 |
+
"learning_rate": 8.708189158016148e-06,
|
| 2018 |
+
"loss": 2.1354,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.491349480968858,
|
| 2023 |
+
"grad_norm": 15.07697868347168,
|
| 2024 |
+
"learning_rate": 8.564013840830449e-06,
|
| 2025 |
+
"loss": 2.1308,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.5,
|
| 2030 |
+
"grad_norm": 13.866670608520508,
|
| 2031 |
+
"learning_rate": 8.419838523644753e-06,
|
| 2032 |
+
"loss": 2.1775,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.508650519031142,
|
| 2037 |
+
"grad_norm": 13.32663631439209,
|
| 2038 |
+
"learning_rate": 8.275663206459055e-06,
|
| 2039 |
+
"loss": 1.7741,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.517301038062284,
|
| 2044 |
+
"grad_norm": 0.0,
|
| 2045 |
+
"learning_rate": 8.131487889273357e-06,
|
| 2046 |
+
"loss": 1.8375,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.5259515570934257,
|
| 2051 |
+
"grad_norm": 8.219161987304688,
|
| 2052 |
+
"learning_rate": 7.987312572087659e-06,
|
| 2053 |
+
"loss": 2.088,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.5346020761245676,
|
| 2058 |
+
"grad_norm": 6.607839107513428,
|
| 2059 |
+
"learning_rate": 7.84313725490196e-06,
|
| 2060 |
+
"loss": 2.2074,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.5432525951557095,
|
| 2065 |
+
"grad_norm": 5.926668643951416,
|
| 2066 |
+
"learning_rate": 7.698961937716263e-06,
|
| 2067 |
+
"loss": 2.0621,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.5519031141868513,
|
| 2072 |
+
"grad_norm": 14.596614837646484,
|
| 2073 |
+
"learning_rate": 7.554786620530566e-06,
|
| 2074 |
+
"loss": 2.1123,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.5605536332179932,
|
| 2079 |
+
"grad_norm": 5.22907829284668,
|
| 2080 |
+
"learning_rate": 7.410611303344868e-06,
|
| 2081 |
+
"loss": 2.1252,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.569204152249135,
|
| 2086 |
+
"grad_norm": 9.819774627685547,
|
| 2087 |
+
"learning_rate": 7.2664359861591705e-06,
|
| 2088 |
+
"loss": 2.069,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.577854671280277,
|
| 2093 |
+
"grad_norm": 7.364692211151123,
|
| 2094 |
+
"learning_rate": 7.1222606689734725e-06,
|
| 2095 |
+
"loss": 2.2576,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.586505190311419,
|
| 2100 |
+
"grad_norm": 7.494932651519775,
|
| 2101 |
+
"learning_rate": 6.978085351787775e-06,
|
| 2102 |
+
"loss": 2.2005,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.595155709342561,
|
| 2107 |
+
"grad_norm": 7.424341678619385,
|
| 2108 |
+
"learning_rate": 6.8339100346020775e-06,
|
| 2109 |
+
"loss": 1.9901,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.6038062283737027,
|
| 2114 |
+
"grad_norm": 7.437248706817627,
|
| 2115 |
+
"learning_rate": 6.689734717416378e-06,
|
| 2116 |
+
"loss": 2.1721,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.612456747404844,
|
| 2121 |
+
"grad_norm": 9.077981948852539,
|
| 2122 |
+
"learning_rate": 6.54555940023068e-06,
|
| 2123 |
+
"loss": 2.2581,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.621107266435986,
|
| 2128 |
+
"grad_norm": 13.281733512878418,
|
| 2129 |
+
"learning_rate": 6.401384083044983e-06,
|
| 2130 |
+
"loss": 2.2352,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.629757785467128,
|
| 2135 |
+
"grad_norm": 7.700297832489014,
|
| 2136 |
+
"learning_rate": 6.257208765859285e-06,
|
| 2137 |
+
"loss": 2.0725,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.63840830449827,
|
| 2142 |
+
"grad_norm": 10.37764835357666,
|
| 2143 |
+
"learning_rate": 6.113033448673587e-06,
|
| 2144 |
+
"loss": 2.0559,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.6470588235294117,
|
| 2149 |
+
"grad_norm": 7.580658912658691,
|
| 2150 |
+
"learning_rate": 5.96885813148789e-06,
|
| 2151 |
+
"loss": 2.103,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.6557093425605536,
|
| 2156 |
+
"grad_norm": 6.464138984680176,
|
| 2157 |
+
"learning_rate": 5.824682814302192e-06,
|
| 2158 |
+
"loss": 1.7107,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.6643598615916955,
|
| 2163 |
+
"grad_norm": 9.265501022338867,
|
| 2164 |
+
"learning_rate": 5.680507497116494e-06,
|
| 2165 |
+
"loss": 2.2707,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.6730103806228374,
|
| 2170 |
+
"grad_norm": 13.189889907836914,
|
| 2171 |
+
"learning_rate": 5.536332179930797e-06,
|
| 2172 |
+
"loss": 2.1762,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.6816608996539792,
|
| 2177 |
+
"grad_norm": 7.203210830688477,
|
| 2178 |
+
"learning_rate": 5.392156862745099e-06,
|
| 2179 |
+
"loss": 2.2239,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.690311418685121,
|
| 2184 |
+
"grad_norm": 4.9819560050964355,
|
| 2185 |
+
"learning_rate": 5.2479815455594e-06,
|
| 2186 |
+
"loss": 2.1301,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.698961937716263,
|
| 2191 |
+
"grad_norm": 9.288585662841797,
|
| 2192 |
+
"learning_rate": 5.103806228373703e-06,
|
| 2193 |
+
"loss": 2.062,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.707612456747405,
|
| 2198 |
+
"grad_norm": 0.0,
|
| 2199 |
+
"learning_rate": 4.959630911188005e-06,
|
| 2200 |
+
"loss": 2.1536,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.716262975778547,
|
| 2205 |
+
"grad_norm": 6.519169807434082,
|
| 2206 |
+
"learning_rate": 4.815455594002307e-06,
|
| 2207 |
+
"loss": 2.1112,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.7249134948096887,
|
| 2212 |
+
"grad_norm": 6.174961566925049,
|
| 2213 |
+
"learning_rate": 4.67128027681661e-06,
|
| 2214 |
+
"loss": 2.2762,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.7335640138408306,
|
| 2219 |
+
"grad_norm": 7.6661553382873535,
|
| 2220 |
+
"learning_rate": 4.527104959630912e-06,
|
| 2221 |
+
"loss": 2.2072,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.7422145328719725,
|
| 2226 |
+
"grad_norm": 13.350886344909668,
|
| 2227 |
+
"learning_rate": 4.382929642445214e-06,
|
| 2228 |
+
"loss": 2.0755,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.750865051903114,
|
| 2233 |
+
"grad_norm": 12.666411399841309,
|
| 2234 |
+
"learning_rate": 4.238754325259516e-06,
|
| 2235 |
+
"loss": 2.2341,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.759515570934256,
|
| 2240 |
+
"grad_norm": 7.005519390106201,
|
| 2241 |
+
"learning_rate": 4.094579008073818e-06,
|
| 2242 |
+
"loss": 1.9337,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.7681660899653977,
|
| 2247 |
+
"grad_norm": 9.67486572265625,
|
| 2248 |
+
"learning_rate": 3.95040369088812e-06,
|
| 2249 |
+
"loss": 2.1086,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.7768166089965396,
|
| 2254 |
+
"grad_norm": 7.333971977233887,
|
| 2255 |
+
"learning_rate": 3.8062283737024224e-06,
|
| 2256 |
+
"loss": 2.227,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.7854671280276815,
|
| 2261 |
+
"grad_norm": 7.098743915557861,
|
| 2262 |
+
"learning_rate": 3.662053056516725e-06,
|
| 2263 |
+
"loss": 1.983,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.7941176470588234,
|
| 2268 |
+
"grad_norm": 6.220783233642578,
|
| 2269 |
+
"learning_rate": 3.517877739331027e-06,
|
| 2270 |
+
"loss": 2.0352,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.8027681660899653,
|
| 2275 |
+
"grad_norm": 4.763909816741943,
|
| 2276 |
+
"learning_rate": 3.3737024221453293e-06,
|
| 2277 |
+
"loss": 2.0064,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.811418685121107,
|
| 2282 |
+
"grad_norm": 7.206717014312744,
|
| 2283 |
+
"learning_rate": 3.229527104959631e-06,
|
| 2284 |
+
"loss": 2.3018,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.820069204152249,
|
| 2289 |
+
"grad_norm": 6.354711055755615,
|
| 2290 |
+
"learning_rate": 3.085351787773933e-06,
|
| 2291 |
+
"loss": 2.0381,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.828719723183391,
|
| 2296 |
+
"grad_norm": 5.708442211151123,
|
| 2297 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 2298 |
+
"loss": 2.027,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.837370242214533,
|
| 2303 |
+
"grad_norm": 6.7003960609436035,
|
| 2304 |
+
"learning_rate": 2.7970011534025375e-06,
|
| 2305 |
+
"loss": 1.9782,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.8460207612456747,
|
| 2310 |
+
"grad_norm": 9.009029388427734,
|
| 2311 |
+
"learning_rate": 2.65282583621684e-06,
|
| 2312 |
+
"loss": 1.8581,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.8546712802768166,
|
| 2317 |
+
"grad_norm": 17.01302719116211,
|
| 2318 |
+
"learning_rate": 2.508650519031142e-06,
|
| 2319 |
+
"loss": 2.1368,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.8633217993079585,
|
| 2324 |
+
"grad_norm": 9.710434913635254,
|
| 2325 |
+
"learning_rate": 2.364475201845444e-06,
|
| 2326 |
+
"loss": 2.1412,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.8719723183391004,
|
| 2331 |
+
"grad_norm": 5.8521552085876465,
|
| 2332 |
+
"learning_rate": 2.2202998846597465e-06,
|
| 2333 |
+
"loss": 1.8587,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.8806228373702423,
|
| 2338 |
+
"grad_norm": 5.487595558166504,
|
| 2339 |
+
"learning_rate": 2.0761245674740485e-06,
|
| 2340 |
+
"loss": 2.3644,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.889273356401384,
|
| 2345 |
+
"grad_norm": 15.505248069763184,
|
| 2346 |
+
"learning_rate": 1.9319492502883506e-06,
|
| 2347 |
+
"loss": 2.4036,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.897923875432526,
|
| 2352 |
+
"grad_norm": 12.557036399841309,
|
| 2353 |
+
"learning_rate": 1.787773933102653e-06,
|
| 2354 |
+
"loss": 2.117,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.906574394463668,
|
| 2359 |
+
"grad_norm": 6.89960241317749,
|
| 2360 |
+
"learning_rate": 1.6435986159169549e-06,
|
| 2361 |
+
"loss": 1.7789,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.91522491349481,
|
| 2366 |
+
"grad_norm": 16.061315536499023,
|
| 2367 |
+
"learning_rate": 1.4994232987312573e-06,
|
| 2368 |
+
"loss": 2.2278,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.9238754325259517,
|
| 2373 |
+
"grad_norm": 7.849757194519043,
|
| 2374 |
+
"learning_rate": 1.3552479815455596e-06,
|
| 2375 |
+
"loss": 2.0531,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.9325259515570936,
|
| 2380 |
+
"grad_norm": 6.261150360107422,
|
| 2381 |
+
"learning_rate": 1.2110726643598616e-06,
|
| 2382 |
+
"loss": 2.2677,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.9411764705882355,
|
| 2387 |
+
"grad_norm": 8.204355239868164,
|
| 2388 |
+
"learning_rate": 1.0668973471741639e-06,
|
| 2389 |
+
"loss": 2.1763,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.9498269896193774,
|
| 2394 |
+
"grad_norm": 7.603878021240234,
|
| 2395 |
+
"learning_rate": 9.22722029988466e-07,
|
| 2396 |
+
"loss": 2.1721,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.9584775086505193,
|
| 2401 |
+
"grad_norm": 10.588805198669434,
|
| 2402 |
+
"learning_rate": 7.785467128027681e-07,
|
| 2403 |
+
"loss": 2.1428,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.967128027681661,
|
| 2408 |
+
"grad_norm": 6.211885452270508,
|
| 2409 |
+
"learning_rate": 6.343713956170704e-07,
|
| 2410 |
+
"loss": 2.2216,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.9757785467128026,
|
| 2415 |
+
"grad_norm": 10.20269775390625,
|
| 2416 |
+
"learning_rate": 4.901960784313725e-07,
|
| 2417 |
+
"loss": 2.0611,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.9844290657439445,
|
| 2422 |
+
"grad_norm": 5.556583881378174,
|
| 2423 |
+
"learning_rate": 3.460207612456748e-07,
|
| 2424 |
+
"loss": 2.3122,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.9930795847750864,
|
| 2429 |
+
"grad_norm": 15.4011812210083,
|
| 2430 |
+
"learning_rate": 2.0184544405997694e-07,
|
| 2431 |
+
"loss": 2.142,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
}
|
| 2434 |
+
],
|
| 2435 |
+
"logging_steps": 10,
|
| 2436 |
+
"max_steps": 3468,
|
| 2437 |
+
"num_input_tokens_seen": 0,
|
| 2438 |
+
"num_train_epochs": 3,
|
| 2439 |
+
"save_steps": 500,
|
| 2440 |
+
"stateful_callbacks": {
|
| 2441 |
+
"TrainerControl": {
|
| 2442 |
+
"args": {
|
| 2443 |
+
"should_epoch_stop": false,
|
| 2444 |
+
"should_evaluate": false,
|
| 2445 |
+
"should_log": false,
|
| 2446 |
+
"should_save": true,
|
| 2447 |
+
"should_training_stop": true
|
| 2448 |
+
},
|
| 2449 |
+
"attributes": {}
|
| 2450 |
+
}
|
| 2451 |
+
},
|
| 2452 |
+
"total_flos": 3623859191808000.0,
|
| 2453 |
+
"train_batch_size": 4,
|
| 2454 |
+
"trial_name": null,
|
| 2455 |
+
"trial_params": null
|
| 2456 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:621b5ab47fda0b3e8cb88d3dccf28d1818147209b376b2925a13c70683978d8e
|
| 3 |
+
size 5368
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|