Upload 13 files
Browse files- config.json +40 -0
- generation_config.json +9 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +37 -0
- tokenizer.json +0 -0
- tokenizer_config.json +64 -0
- trainer_state.json +2036 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"LABEL_0": 0
|
| 18 |
+
},
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"model_type": "gpt2",
|
| 21 |
+
"n_ctx": 2048,
|
| 22 |
+
"n_embd": 768,
|
| 23 |
+
"n_head": 12,
|
| 24 |
+
"n_inner": null,
|
| 25 |
+
"n_layer": 12,
|
| 26 |
+
"n_positions": 2048,
|
| 27 |
+
"pad_token_id": 2,
|
| 28 |
+
"reorder_and_upcast_attn": false,
|
| 29 |
+
"resid_pdrop": 0.1,
|
| 30 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 31 |
+
"scale_attn_weights": true,
|
| 32 |
+
"summary_activation": null,
|
| 33 |
+
"summary_first_dropout": 0.1,
|
| 34 |
+
"summary_proj_to_labels": true,
|
| 35 |
+
"summary_type": "cls_index",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"transformers_version": "4.57.3",
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 50257
|
| 40 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
2
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 2,
|
| 8 |
+
"transformers_version": "4.57.3"
|
| 9 |
+
}
|
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 sha256:941af006c53b04c8b3d79d6a92c9d95357ad1a9d1f0e35a3fe441bd7b3025b7b
|
| 3 |
+
size 500919936
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9b7c46511f5499ee004484bc8087636f995b11dfdc79096308d94a3e514e9a87
|
| 3 |
+
size 1001933754
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b134f8e3af884ecbd7d1e2ca01867adadd87959eaa2547534fb7d1ad32f8a29
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d90d45eba9611e74e6ec237c3541e68dbb441dd6187112cfd8db9bbad5d0e7af
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"mask_token": {
|
| 17 |
+
"content": "<mask>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"pad_token": {
|
| 24 |
+
"content": "</s>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "<unk>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<pad>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": true,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": true,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"3": {
|
| 30 |
+
"content": "<unk>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": true,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"4": {
|
| 38 |
+
"content": "<mask>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
}
|
| 45 |
+
},
|
| 46 |
+
"bos_token": "<s>",
|
| 47 |
+
"clean_up_tokenization_spaces": true,
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"max_length": 128,
|
| 53 |
+
"model_max_length": 2048,
|
| 54 |
+
"pad_to_multiple_of": null,
|
| 55 |
+
"pad_token": "</s>",
|
| 56 |
+
"pad_token_type_id": 0,
|
| 57 |
+
"padding_side": "left",
|
| 58 |
+
"stride": 0,
|
| 59 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 60 |
+
"truncation_side": "left",
|
| 61 |
+
"truncation_strategy": "longest_first",
|
| 62 |
+
"trust_remote_code": false,
|
| 63 |
+
"unk_token": "<unk>"
|
| 64 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2036 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 28604,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0034960145434205004,
|
| 14 |
+
"grad_norm": 3.56770658493042,
|
| 15 |
+
"learning_rate": 4.982694728010069e-05,
|
| 16 |
+
"loss": 5.1215,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.006992029086841001,
|
| 21 |
+
"grad_norm": 3.0420525074005127,
|
| 22 |
+
"learning_rate": 4.965214655292966e-05,
|
| 23 |
+
"loss": 5.1137,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.010488043630261502,
|
| 28 |
+
"grad_norm": 3.0995962619781494,
|
| 29 |
+
"learning_rate": 4.9477345825758635e-05,
|
| 30 |
+
"loss": 5.1482,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.013984058173682002,
|
| 35 |
+
"grad_norm": 2.8879716396331787,
|
| 36 |
+
"learning_rate": 4.930254509858761e-05,
|
| 37 |
+
"loss": 5.1249,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0174800727171025,
|
| 42 |
+
"grad_norm": 2.7549917697906494,
|
| 43 |
+
"learning_rate": 4.912774437141659e-05,
|
| 44 |
+
"loss": 5.0889,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.020976087260523003,
|
| 49 |
+
"grad_norm": 2.814000129699707,
|
| 50 |
+
"learning_rate": 4.895294364424556e-05,
|
| 51 |
+
"loss": 5.1358,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.024472101803943505,
|
| 56 |
+
"grad_norm": 2.6768572330474854,
|
| 57 |
+
"learning_rate": 4.877814291707454e-05,
|
| 58 |
+
"loss": 5.1105,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.027968116347364003,
|
| 63 |
+
"grad_norm": 2.678793430328369,
|
| 64 |
+
"learning_rate": 4.860334218990351e-05,
|
| 65 |
+
"loss": 5.1058,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.03146413089078451,
|
| 70 |
+
"grad_norm": 3.129312753677368,
|
| 71 |
+
"learning_rate": 4.8428541462732484e-05,
|
| 72 |
+
"loss": 5.1086,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.034960145434205,
|
| 77 |
+
"grad_norm": 2.9247419834136963,
|
| 78 |
+
"learning_rate": 4.8253740735561464e-05,
|
| 79 |
+
"loss": 5.1296,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.038456159977625505,
|
| 84 |
+
"grad_norm": 2.542954206466675,
|
| 85 |
+
"learning_rate": 4.807894000839044e-05,
|
| 86 |
+
"loss": 5.0857,
|
| 87 |
+
"step": 1100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04195217452104601,
|
| 91 |
+
"grad_norm": 2.79154372215271,
|
| 92 |
+
"learning_rate": 4.790413928121941e-05,
|
| 93 |
+
"loss": 5.1134,
|
| 94 |
+
"step": 1200
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.04544818906446651,
|
| 98 |
+
"grad_norm": 2.6644339561462402,
|
| 99 |
+
"learning_rate": 4.7729338554048386e-05,
|
| 100 |
+
"loss": 5.105,
|
| 101 |
+
"step": 1300
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.04894420360788701,
|
| 105 |
+
"grad_norm": 2.7891204357147217,
|
| 106 |
+
"learning_rate": 4.755453782687736e-05,
|
| 107 |
+
"loss": 5.0809,
|
| 108 |
+
"step": 1400
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.05244021815130751,
|
| 112 |
+
"grad_norm": 2.664785623550415,
|
| 113 |
+
"learning_rate": 4.737973709970634e-05,
|
| 114 |
+
"loss": 5.0933,
|
| 115 |
+
"step": 1500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.05593623269472801,
|
| 119 |
+
"grad_norm": 3.264219284057617,
|
| 120 |
+
"learning_rate": 4.720493637253531e-05,
|
| 121 |
+
"loss": 5.1067,
|
| 122 |
+
"step": 1600
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.05943224723814851,
|
| 126 |
+
"grad_norm": 2.754682779312134,
|
| 127 |
+
"learning_rate": 4.703013564536429e-05,
|
| 128 |
+
"loss": 5.0338,
|
| 129 |
+
"step": 1700
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06292826178156902,
|
| 133 |
+
"grad_norm": 2.821290969848633,
|
| 134 |
+
"learning_rate": 4.685533491819326e-05,
|
| 135 |
+
"loss": 5.0775,
|
| 136 |
+
"step": 1800
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.06642427632498951,
|
| 140 |
+
"grad_norm": 2.995471954345703,
|
| 141 |
+
"learning_rate": 4.6680534191022234e-05,
|
| 142 |
+
"loss": 5.0705,
|
| 143 |
+
"step": 1900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.06992029086841,
|
| 147 |
+
"grad_norm": 2.7494113445281982,
|
| 148 |
+
"learning_rate": 4.6505733463851215e-05,
|
| 149 |
+
"loss": 5.0112,
|
| 150 |
+
"step": 2000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.07341630541183052,
|
| 154 |
+
"grad_norm": 3.065732479095459,
|
| 155 |
+
"learning_rate": 4.633093273668019e-05,
|
| 156 |
+
"loss": 5.1087,
|
| 157 |
+
"step": 2100
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.07691231995525101,
|
| 161 |
+
"grad_norm": 2.742866039276123,
|
| 162 |
+
"learning_rate": 4.615613200950916e-05,
|
| 163 |
+
"loss": 5.0839,
|
| 164 |
+
"step": 2200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08040833449867152,
|
| 168 |
+
"grad_norm": 2.7256627082824707,
|
| 169 |
+
"learning_rate": 4.5981331282338136e-05,
|
| 170 |
+
"loss": 5.0284,
|
| 171 |
+
"step": 2300
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.08390434904209201,
|
| 175 |
+
"grad_norm": 2.7911083698272705,
|
| 176 |
+
"learning_rate": 4.580653055516711e-05,
|
| 177 |
+
"loss": 5.0478,
|
| 178 |
+
"step": 2400
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.08740036358551252,
|
| 182 |
+
"grad_norm": 2.8085787296295166,
|
| 183 |
+
"learning_rate": 4.563172982799609e-05,
|
| 184 |
+
"loss": 5.0503,
|
| 185 |
+
"step": 2500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.09089637812893302,
|
| 189 |
+
"grad_norm": 2.870654344558716,
|
| 190 |
+
"learning_rate": 4.5456929100825064e-05,
|
| 191 |
+
"loss": 5.0605,
|
| 192 |
+
"step": 2600
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.09439239267235351,
|
| 196 |
+
"grad_norm": 2.848191976547241,
|
| 197 |
+
"learning_rate": 4.528212837365404e-05,
|
| 198 |
+
"loss": 5.0471,
|
| 199 |
+
"step": 2700
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.09788840721577402,
|
| 203 |
+
"grad_norm": 2.4951555728912354,
|
| 204 |
+
"learning_rate": 4.510732764648301e-05,
|
| 205 |
+
"loss": 5.0179,
|
| 206 |
+
"step": 2800
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.10138442175919452,
|
| 210 |
+
"grad_norm": 2.7134485244750977,
|
| 211 |
+
"learning_rate": 4.4932526919311985e-05,
|
| 212 |
+
"loss": 5.0412,
|
| 213 |
+
"step": 2900
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.10488043630261502,
|
| 217 |
+
"grad_norm": 3.0360782146453857,
|
| 218 |
+
"learning_rate": 4.4757726192140965e-05,
|
| 219 |
+
"loss": 5.0416,
|
| 220 |
+
"step": 3000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.10837645084603552,
|
| 224 |
+
"grad_norm": 2.8149096965789795,
|
| 225 |
+
"learning_rate": 4.458292546496994e-05,
|
| 226 |
+
"loss": 5.0165,
|
| 227 |
+
"step": 3100
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.11187246538945601,
|
| 231 |
+
"grad_norm": 3.08152174949646,
|
| 232 |
+
"learning_rate": 4.440812473779891e-05,
|
| 233 |
+
"loss": 4.9924,
|
| 234 |
+
"step": 3200
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.11536847993287652,
|
| 238 |
+
"grad_norm": 2.820800542831421,
|
| 239 |
+
"learning_rate": 4.4233324010627886e-05,
|
| 240 |
+
"loss": 5.0404,
|
| 241 |
+
"step": 3300
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.11886449447629702,
|
| 245 |
+
"grad_norm": 2.6280956268310547,
|
| 246 |
+
"learning_rate": 4.405852328345686e-05,
|
| 247 |
+
"loss": 5.0052,
|
| 248 |
+
"step": 3400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.12236050901971753,
|
| 252 |
+
"grad_norm": 2.5183768272399902,
|
| 253 |
+
"learning_rate": 4.388372255628584e-05,
|
| 254 |
+
"loss": 5.0197,
|
| 255 |
+
"step": 3500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.12585652356313803,
|
| 259 |
+
"grad_norm": 2.802274465560913,
|
| 260 |
+
"learning_rate": 4.3708921829114814e-05,
|
| 261 |
+
"loss": 4.99,
|
| 262 |
+
"step": 3600
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.12935253810655853,
|
| 266 |
+
"grad_norm": 2.8082516193389893,
|
| 267 |
+
"learning_rate": 4.353412110194379e-05,
|
| 268 |
+
"loss": 4.998,
|
| 269 |
+
"step": 3700
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.13284855264997902,
|
| 273 |
+
"grad_norm": 2.8567910194396973,
|
| 274 |
+
"learning_rate": 4.335932037477276e-05,
|
| 275 |
+
"loss": 5.007,
|
| 276 |
+
"step": 3800
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.13634456719339952,
|
| 280 |
+
"grad_norm": 2.6747946739196777,
|
| 281 |
+
"learning_rate": 4.3184519647601735e-05,
|
| 282 |
+
"loss": 5.0372,
|
| 283 |
+
"step": 3900
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.13984058173682,
|
| 287 |
+
"grad_norm": 3.1875665187835693,
|
| 288 |
+
"learning_rate": 4.3009718920430715e-05,
|
| 289 |
+
"loss": 5.008,
|
| 290 |
+
"step": 4000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.14333659628024054,
|
| 294 |
+
"grad_norm": 2.5646297931671143,
|
| 295 |
+
"learning_rate": 4.283491819325969e-05,
|
| 296 |
+
"loss": 5.0014,
|
| 297 |
+
"step": 4100
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.14683261082366103,
|
| 301 |
+
"grad_norm": 2.9062674045562744,
|
| 302 |
+
"learning_rate": 4.266011746608866e-05,
|
| 303 |
+
"loss": 4.999,
|
| 304 |
+
"step": 4200
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.15032862536708153,
|
| 308 |
+
"grad_norm": 2.6203787326812744,
|
| 309 |
+
"learning_rate": 4.2485316738917637e-05,
|
| 310 |
+
"loss": 4.9994,
|
| 311 |
+
"step": 4300
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.15382463991050202,
|
| 315 |
+
"grad_norm": 2.530931234359741,
|
| 316 |
+
"learning_rate": 4.231051601174661e-05,
|
| 317 |
+
"loss": 4.9835,
|
| 318 |
+
"step": 4400
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.15732065445392251,
|
| 322 |
+
"grad_norm": 2.436789035797119,
|
| 323 |
+
"learning_rate": 4.2135715284575584e-05,
|
| 324 |
+
"loss": 4.9711,
|
| 325 |
+
"step": 4500
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.16081666899734304,
|
| 329 |
+
"grad_norm": 2.867676019668579,
|
| 330 |
+
"learning_rate": 4.1960914557404564e-05,
|
| 331 |
+
"loss": 4.9843,
|
| 332 |
+
"step": 4600
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.16431268354076353,
|
| 336 |
+
"grad_norm": 3.0732126235961914,
|
| 337 |
+
"learning_rate": 4.178611383023354e-05,
|
| 338 |
+
"loss": 4.9614,
|
| 339 |
+
"step": 4700
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.16780869808418403,
|
| 343 |
+
"grad_norm": 2.7034592628479004,
|
| 344 |
+
"learning_rate": 4.161131310306251e-05,
|
| 345 |
+
"loss": 5.0208,
|
| 346 |
+
"step": 4800
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.17130471262760452,
|
| 350 |
+
"grad_norm": 2.957083225250244,
|
| 351 |
+
"learning_rate": 4.1436512375891485e-05,
|
| 352 |
+
"loss": 4.9884,
|
| 353 |
+
"step": 4900
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.17480072717102504,
|
| 357 |
+
"grad_norm": 2.7425382137298584,
|
| 358 |
+
"learning_rate": 4.126171164872046e-05,
|
| 359 |
+
"loss": 4.9781,
|
| 360 |
+
"step": 5000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.17829674171444554,
|
| 364 |
+
"grad_norm": 2.872343063354492,
|
| 365 |
+
"learning_rate": 4.108691092154944e-05,
|
| 366 |
+
"loss": 5.0063,
|
| 367 |
+
"step": 5100
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.18179275625786603,
|
| 371 |
+
"grad_norm": 2.698291301727295,
|
| 372 |
+
"learning_rate": 4.091211019437841e-05,
|
| 373 |
+
"loss": 5.0118,
|
| 374 |
+
"step": 5200
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.18528877080128653,
|
| 378 |
+
"grad_norm": 2.5991404056549072,
|
| 379 |
+
"learning_rate": 4.073730946720739e-05,
|
| 380 |
+
"loss": 4.9926,
|
| 381 |
+
"step": 5300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.18878478534470702,
|
| 385 |
+
"grad_norm": 2.681043863296509,
|
| 386 |
+
"learning_rate": 4.056250874003636e-05,
|
| 387 |
+
"loss": 4.9436,
|
| 388 |
+
"step": 5400
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.19228079988812755,
|
| 392 |
+
"grad_norm": 2.5356483459472656,
|
| 393 |
+
"learning_rate": 4.0387708012865334e-05,
|
| 394 |
+
"loss": 4.9587,
|
| 395 |
+
"step": 5500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.19577681443154804,
|
| 399 |
+
"grad_norm": 2.8511860370635986,
|
| 400 |
+
"learning_rate": 4.0212907285694315e-05,
|
| 401 |
+
"loss": 4.9898,
|
| 402 |
+
"step": 5600
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.19927282897496854,
|
| 406 |
+
"grad_norm": 2.5632340908050537,
|
| 407 |
+
"learning_rate": 4.003810655852329e-05,
|
| 408 |
+
"loss": 4.9834,
|
| 409 |
+
"step": 5700
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.20276884351838903,
|
| 413 |
+
"grad_norm": 3.114943742752075,
|
| 414 |
+
"learning_rate": 3.986330583135226e-05,
|
| 415 |
+
"loss": 4.9516,
|
| 416 |
+
"step": 5800
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.20626485806180953,
|
| 420 |
+
"grad_norm": 2.759326457977295,
|
| 421 |
+
"learning_rate": 3.9688505104181236e-05,
|
| 422 |
+
"loss": 5.0019,
|
| 423 |
+
"step": 5900
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.20976087260523005,
|
| 427 |
+
"grad_norm": 2.8129642009735107,
|
| 428 |
+
"learning_rate": 3.951370437701021e-05,
|
| 429 |
+
"loss": 4.9568,
|
| 430 |
+
"step": 6000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.21325688714865054,
|
| 434 |
+
"grad_norm": 2.8741278648376465,
|
| 435 |
+
"learning_rate": 3.933890364983919e-05,
|
| 436 |
+
"loss": 4.9049,
|
| 437 |
+
"step": 6100
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.21675290169207104,
|
| 441 |
+
"grad_norm": 2.5861387252807617,
|
| 442 |
+
"learning_rate": 3.9164102922668164e-05,
|
| 443 |
+
"loss": 5.0054,
|
| 444 |
+
"step": 6200
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.22024891623549153,
|
| 448 |
+
"grad_norm": 2.668247938156128,
|
| 449 |
+
"learning_rate": 3.898930219549714e-05,
|
| 450 |
+
"loss": 4.9488,
|
| 451 |
+
"step": 6300
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.22374493077891203,
|
| 455 |
+
"grad_norm": 2.8038411140441895,
|
| 456 |
+
"learning_rate": 3.881450146832611e-05,
|
| 457 |
+
"loss": 4.9846,
|
| 458 |
+
"step": 6400
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.22724094532233255,
|
| 462 |
+
"grad_norm": 2.64792799949646,
|
| 463 |
+
"learning_rate": 3.8639700741155085e-05,
|
| 464 |
+
"loss": 4.9734,
|
| 465 |
+
"step": 6500
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.23073695986575304,
|
| 469 |
+
"grad_norm": 3.3240697383880615,
|
| 470 |
+
"learning_rate": 3.8464900013984065e-05,
|
| 471 |
+
"loss": 4.9348,
|
| 472 |
+
"step": 6600
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.23423297440917354,
|
| 476 |
+
"grad_norm": 2.762397050857544,
|
| 477 |
+
"learning_rate": 3.829009928681304e-05,
|
| 478 |
+
"loss": 4.9853,
|
| 479 |
+
"step": 6700
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.23772898895259403,
|
| 483 |
+
"grad_norm": 2.5263381004333496,
|
| 484 |
+
"learning_rate": 3.811529855964201e-05,
|
| 485 |
+
"loss": 4.9199,
|
| 486 |
+
"step": 6800
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.24122500349601456,
|
| 490 |
+
"grad_norm": 3.1513848304748535,
|
| 491 |
+
"learning_rate": 3.7940497832470986e-05,
|
| 492 |
+
"loss": 4.9958,
|
| 493 |
+
"step": 6900
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.24472101803943505,
|
| 497 |
+
"grad_norm": 2.661020278930664,
|
| 498 |
+
"learning_rate": 3.776569710529996e-05,
|
| 499 |
+
"loss": 4.9228,
|
| 500 |
+
"step": 7000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.24821703258285555,
|
| 504 |
+
"grad_norm": 2.742587089538574,
|
| 505 |
+
"learning_rate": 3.759089637812894e-05,
|
| 506 |
+
"loss": 4.9588,
|
| 507 |
+
"step": 7100
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.25171304712627607,
|
| 511 |
+
"grad_norm": 2.8065550327301025,
|
| 512 |
+
"learning_rate": 3.7416095650957914e-05,
|
| 513 |
+
"loss": 4.9245,
|
| 514 |
+
"step": 7200
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.25520906166969654,
|
| 518 |
+
"grad_norm": 2.639291524887085,
|
| 519 |
+
"learning_rate": 3.724129492378689e-05,
|
| 520 |
+
"loss": 4.9722,
|
| 521 |
+
"step": 7300
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.25870507621311706,
|
| 525 |
+
"grad_norm": 2.7371039390563965,
|
| 526 |
+
"learning_rate": 3.706649419661586e-05,
|
| 527 |
+
"loss": 4.9584,
|
| 528 |
+
"step": 7400
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.2622010907565375,
|
| 532 |
+
"grad_norm": 2.891352891921997,
|
| 533 |
+
"learning_rate": 3.6891693469444835e-05,
|
| 534 |
+
"loss": 4.939,
|
| 535 |
+
"step": 7500
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.26569710529995805,
|
| 539 |
+
"grad_norm": 2.5609800815582275,
|
| 540 |
+
"learning_rate": 3.671689274227381e-05,
|
| 541 |
+
"loss": 4.9424,
|
| 542 |
+
"step": 7600
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.26919311984337857,
|
| 546 |
+
"grad_norm": 2.6541666984558105,
|
| 547 |
+
"learning_rate": 3.654209201510279e-05,
|
| 548 |
+
"loss": 4.9347,
|
| 549 |
+
"step": 7700
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.27268913438679904,
|
| 553 |
+
"grad_norm": 3.461524724960327,
|
| 554 |
+
"learning_rate": 3.636729128793176e-05,
|
| 555 |
+
"loss": 4.9917,
|
| 556 |
+
"step": 7800
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.27618514893021956,
|
| 560 |
+
"grad_norm": 2.6681628227233887,
|
| 561 |
+
"learning_rate": 3.6192490560760736e-05,
|
| 562 |
+
"loss": 4.9667,
|
| 563 |
+
"step": 7900
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.27968116347364,
|
| 567 |
+
"grad_norm": 2.8997554779052734,
|
| 568 |
+
"learning_rate": 3.601768983358971e-05,
|
| 569 |
+
"loss": 4.8909,
|
| 570 |
+
"step": 8000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.28317717801706055,
|
| 574 |
+
"grad_norm": 2.552457094192505,
|
| 575 |
+
"learning_rate": 3.5842889106418684e-05,
|
| 576 |
+
"loss": 4.9577,
|
| 577 |
+
"step": 8100
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.28667319256048107,
|
| 581 |
+
"grad_norm": 2.6073288917541504,
|
| 582 |
+
"learning_rate": 3.566808837924766e-05,
|
| 583 |
+
"loss": 4.9378,
|
| 584 |
+
"step": 8200
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.29016920710390154,
|
| 588 |
+
"grad_norm": 2.604503870010376,
|
| 589 |
+
"learning_rate": 3.549328765207663e-05,
|
| 590 |
+
"loss": 4.9283,
|
| 591 |
+
"step": 8300
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.29366522164732206,
|
| 595 |
+
"grad_norm": 2.9705913066864014,
|
| 596 |
+
"learning_rate": 3.531848692490561e-05,
|
| 597 |
+
"loss": 4.9143,
|
| 598 |
+
"step": 8400
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.29716123619074253,
|
| 602 |
+
"grad_norm": 2.5234224796295166,
|
| 603 |
+
"learning_rate": 3.5143686197734585e-05,
|
| 604 |
+
"loss": 4.9231,
|
| 605 |
+
"step": 8500
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.30065725073416305,
|
| 609 |
+
"grad_norm": 2.8847386837005615,
|
| 610 |
+
"learning_rate": 3.496888547056356e-05,
|
| 611 |
+
"loss": 4.9186,
|
| 612 |
+
"step": 8600
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3041532652775836,
|
| 616 |
+
"grad_norm": 3.093806505203247,
|
| 617 |
+
"learning_rate": 3.479408474339253e-05,
|
| 618 |
+
"loss": 4.895,
|
| 619 |
+
"step": 8700
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.30764927982100404,
|
| 623 |
+
"grad_norm": 3.003735303878784,
|
| 624 |
+
"learning_rate": 3.4619284016221506e-05,
|
| 625 |
+
"loss": 4.8982,
|
| 626 |
+
"step": 8800
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.31114529436442456,
|
| 630 |
+
"grad_norm": 2.972370147705078,
|
| 631 |
+
"learning_rate": 3.444448328905048e-05,
|
| 632 |
+
"loss": 4.9627,
|
| 633 |
+
"step": 8900
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.31464130890784503,
|
| 637 |
+
"grad_norm": 2.95778751373291,
|
| 638 |
+
"learning_rate": 3.426968256187946e-05,
|
| 639 |
+
"loss": 4.8926,
|
| 640 |
+
"step": 9000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.31813732345126555,
|
| 644 |
+
"grad_norm": 2.7067129611968994,
|
| 645 |
+
"learning_rate": 3.4094881834708434e-05,
|
| 646 |
+
"loss": 4.9371,
|
| 647 |
+
"step": 9100
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3216333379946861,
|
| 651 |
+
"grad_norm": 2.6828505992889404,
|
| 652 |
+
"learning_rate": 3.392008110753741e-05,
|
| 653 |
+
"loss": 4.9095,
|
| 654 |
+
"step": 9200
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.32512935253810654,
|
| 658 |
+
"grad_norm": 2.7085490226745605,
|
| 659 |
+
"learning_rate": 3.374528038036638e-05,
|
| 660 |
+
"loss": 4.895,
|
| 661 |
+
"step": 9300
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.32862536708152706,
|
| 665 |
+
"grad_norm": 2.601956605911255,
|
| 666 |
+
"learning_rate": 3.3570479653195355e-05,
|
| 667 |
+
"loss": 4.9177,
|
| 668 |
+
"step": 9400
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.3321213816249476,
|
| 672 |
+
"grad_norm": 2.70379900932312,
|
| 673 |
+
"learning_rate": 3.339567892602433e-05,
|
| 674 |
+
"loss": 4.9543,
|
| 675 |
+
"step": 9500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.33561739616836805,
|
| 679 |
+
"grad_norm": 2.8263227939605713,
|
| 680 |
+
"learning_rate": 3.322087819885331e-05,
|
| 681 |
+
"loss": 4.9403,
|
| 682 |
+
"step": 9600
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.3391134107117886,
|
| 686 |
+
"grad_norm": 2.645223617553711,
|
| 687 |
+
"learning_rate": 3.304607747168228e-05,
|
| 688 |
+
"loss": 4.9174,
|
| 689 |
+
"step": 9700
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.34260942525520904,
|
| 693 |
+
"grad_norm": 3.0595059394836426,
|
| 694 |
+
"learning_rate": 3.2871276744511257e-05,
|
| 695 |
+
"loss": 4.9093,
|
| 696 |
+
"step": 9800
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.34610543979862957,
|
| 700 |
+
"grad_norm": 3.042128324508667,
|
| 701 |
+
"learning_rate": 3.269647601734023e-05,
|
| 702 |
+
"loss": 4.8524,
|
| 703 |
+
"step": 9900
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3496014543420501,
|
| 707 |
+
"grad_norm": 2.5243098735809326,
|
| 708 |
+
"learning_rate": 3.2521675290169204e-05,
|
| 709 |
+
"loss": 4.8915,
|
| 710 |
+
"step": 10000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.35309746888547056,
|
| 714 |
+
"grad_norm": 2.696924924850464,
|
| 715 |
+
"learning_rate": 3.2346874562998184e-05,
|
| 716 |
+
"loss": 4.8823,
|
| 717 |
+
"step": 10100
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.3565934834288911,
|
| 721 |
+
"grad_norm": 3.047484874725342,
|
| 722 |
+
"learning_rate": 3.217207383582716e-05,
|
| 723 |
+
"loss": 4.8717,
|
| 724 |
+
"step": 10200
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.36008949797231155,
|
| 728 |
+
"grad_norm": 2.573148250579834,
|
| 729 |
+
"learning_rate": 3.199727310865613e-05,
|
| 730 |
+
"loss": 4.9094,
|
| 731 |
+
"step": 10300
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.36358551251573207,
|
| 735 |
+
"grad_norm": 2.771249532699585,
|
| 736 |
+
"learning_rate": 3.1822472381485105e-05,
|
| 737 |
+
"loss": 4.8894,
|
| 738 |
+
"step": 10400
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.3670815270591526,
|
| 742 |
+
"grad_norm": 2.7111799716949463,
|
| 743 |
+
"learning_rate": 3.164767165431408e-05,
|
| 744 |
+
"loss": 4.8781,
|
| 745 |
+
"step": 10500
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.37057754160257306,
|
| 749 |
+
"grad_norm": 2.443671941757202,
|
| 750 |
+
"learning_rate": 3.147287092714306e-05,
|
| 751 |
+
"loss": 4.911,
|
| 752 |
+
"step": 10600
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.3740735561459936,
|
| 756 |
+
"grad_norm": 3.0202887058258057,
|
| 757 |
+
"learning_rate": 3.129807019997203e-05,
|
| 758 |
+
"loss": 4.8911,
|
| 759 |
+
"step": 10700
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.37756957068941405,
|
| 763 |
+
"grad_norm": 2.860243320465088,
|
| 764 |
+
"learning_rate": 3.112326947280101e-05,
|
| 765 |
+
"loss": 4.8759,
|
| 766 |
+
"step": 10800
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.38106558523283457,
|
| 770 |
+
"grad_norm": 2.7219772338867188,
|
| 771 |
+
"learning_rate": 3.094846874562998e-05,
|
| 772 |
+
"loss": 4.8958,
|
| 773 |
+
"step": 10900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.3845615997762551,
|
| 777 |
+
"grad_norm": 2.7428410053253174,
|
| 778 |
+
"learning_rate": 3.0773668018458954e-05,
|
| 779 |
+
"loss": 4.8559,
|
| 780 |
+
"step": 11000
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.38805761431967556,
|
| 784 |
+
"grad_norm": 2.6882712841033936,
|
| 785 |
+
"learning_rate": 3.0598867291287935e-05,
|
| 786 |
+
"loss": 4.9264,
|
| 787 |
+
"step": 11100
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.3915536288630961,
|
| 791 |
+
"grad_norm": 2.3838911056518555,
|
| 792 |
+
"learning_rate": 3.042406656411691e-05,
|
| 793 |
+
"loss": 4.8896,
|
| 794 |
+
"step": 11200
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.39504964340651655,
|
| 798 |
+
"grad_norm": 2.679856538772583,
|
| 799 |
+
"learning_rate": 3.0249265836945882e-05,
|
| 800 |
+
"loss": 4.8963,
|
| 801 |
+
"step": 11300
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.39854565794993707,
|
| 805 |
+
"grad_norm": 2.4506402015686035,
|
| 806 |
+
"learning_rate": 3.007446510977486e-05,
|
| 807 |
+
"loss": 4.917,
|
| 808 |
+
"step": 11400
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4020416724933576,
|
| 812 |
+
"grad_norm": 3.169481039047241,
|
| 813 |
+
"learning_rate": 2.9899664382603833e-05,
|
| 814 |
+
"loss": 4.8908,
|
| 815 |
+
"step": 11500
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.40553768703677806,
|
| 819 |
+
"grad_norm": 2.744379997253418,
|
| 820 |
+
"learning_rate": 2.9724863655432806e-05,
|
| 821 |
+
"loss": 4.8491,
|
| 822 |
+
"step": 11600
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.4090337015801986,
|
| 826 |
+
"grad_norm": 2.633491039276123,
|
| 827 |
+
"learning_rate": 2.9550062928261784e-05,
|
| 828 |
+
"loss": 4.8553,
|
| 829 |
+
"step": 11700
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.41252971612361905,
|
| 833 |
+
"grad_norm": 2.816256284713745,
|
| 834 |
+
"learning_rate": 2.9375262201090757e-05,
|
| 835 |
+
"loss": 4.8653,
|
| 836 |
+
"step": 11800
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4160257306670396,
|
| 840 |
+
"grad_norm": 2.6040494441986084,
|
| 841 |
+
"learning_rate": 2.920046147391973e-05,
|
| 842 |
+
"loss": 4.8381,
|
| 843 |
+
"step": 11900
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.4195217452104601,
|
| 847 |
+
"grad_norm": 3.032259225845337,
|
| 848 |
+
"learning_rate": 2.9025660746748705e-05,
|
| 849 |
+
"loss": 4.912,
|
| 850 |
+
"step": 12000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.42301775975388056,
|
| 854 |
+
"grad_norm": 3.134244203567505,
|
| 855 |
+
"learning_rate": 2.885086001957768e-05,
|
| 856 |
+
"loss": 4.8971,
|
| 857 |
+
"step": 12100
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.4265137742973011,
|
| 861 |
+
"grad_norm": 2.9555444717407227,
|
| 862 |
+
"learning_rate": 2.867605929240666e-05,
|
| 863 |
+
"loss": 4.8879,
|
| 864 |
+
"step": 12200
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.43000978884072155,
|
| 868 |
+
"grad_norm": 2.4772422313690186,
|
| 869 |
+
"learning_rate": 2.8501258565235632e-05,
|
| 870 |
+
"loss": 4.8624,
|
| 871 |
+
"step": 12300
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.4335058033841421,
|
| 875 |
+
"grad_norm": 2.8734424114227295,
|
| 876 |
+
"learning_rate": 2.8326457838064606e-05,
|
| 877 |
+
"loss": 4.8553,
|
| 878 |
+
"step": 12400
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.4370018179275626,
|
| 882 |
+
"grad_norm": 2.7253541946411133,
|
| 883 |
+
"learning_rate": 2.815165711089358e-05,
|
| 884 |
+
"loss": 4.8841,
|
| 885 |
+
"step": 12500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.44049783247098306,
|
| 889 |
+
"grad_norm": 3.075535297393799,
|
| 890 |
+
"learning_rate": 2.7976856383722553e-05,
|
| 891 |
+
"loss": 4.8871,
|
| 892 |
+
"step": 12600
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.4439938470144036,
|
| 896 |
+
"grad_norm": 2.6050570011138916,
|
| 897 |
+
"learning_rate": 2.7802055656551534e-05,
|
| 898 |
+
"loss": 4.8382,
|
| 899 |
+
"step": 12700
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.44748986155782405,
|
| 903 |
+
"grad_norm": 2.6890673637390137,
|
| 904 |
+
"learning_rate": 2.7627254929380508e-05,
|
| 905 |
+
"loss": 4.9089,
|
| 906 |
+
"step": 12800
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.4509858761012446,
|
| 910 |
+
"grad_norm": 2.602142572402954,
|
| 911 |
+
"learning_rate": 2.745245420220948e-05,
|
| 912 |
+
"loss": 4.8587,
|
| 913 |
+
"step": 12900
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.4544818906446651,
|
| 917 |
+
"grad_norm": 2.8939311504364014,
|
| 918 |
+
"learning_rate": 2.7277653475038455e-05,
|
| 919 |
+
"loss": 4.8508,
|
| 920 |
+
"step": 13000
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.45797790518808557,
|
| 924 |
+
"grad_norm": 3.0450758934020996,
|
| 925 |
+
"learning_rate": 2.710285274786743e-05,
|
| 926 |
+
"loss": 4.8884,
|
| 927 |
+
"step": 13100
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.4614739197315061,
|
| 931 |
+
"grad_norm": 2.4321885108947754,
|
| 932 |
+
"learning_rate": 2.692805202069641e-05,
|
| 933 |
+
"loss": 4.8996,
|
| 934 |
+
"step": 13200
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.4649699342749266,
|
| 938 |
+
"grad_norm": 2.4929943084716797,
|
| 939 |
+
"learning_rate": 2.6753251293525383e-05,
|
| 940 |
+
"loss": 4.8747,
|
| 941 |
+
"step": 13300
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.4684659488183471,
|
| 945 |
+
"grad_norm": 2.8290367126464844,
|
| 946 |
+
"learning_rate": 2.6578450566354356e-05,
|
| 947 |
+
"loss": 4.9006,
|
| 948 |
+
"step": 13400
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.4719619633617676,
|
| 952 |
+
"grad_norm": 3.063918352127075,
|
| 953 |
+
"learning_rate": 2.640364983918333e-05,
|
| 954 |
+
"loss": 4.8664,
|
| 955 |
+
"step": 13500
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.47545797790518807,
|
| 959 |
+
"grad_norm": 3.0412003993988037,
|
| 960 |
+
"learning_rate": 2.6228849112012304e-05,
|
| 961 |
+
"loss": 4.8873,
|
| 962 |
+
"step": 13600
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.4789539924486086,
|
| 966 |
+
"grad_norm": 3.148043155670166,
|
| 967 |
+
"learning_rate": 2.6054048384841284e-05,
|
| 968 |
+
"loss": 4.8698,
|
| 969 |
+
"step": 13700
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.4824500069920291,
|
| 973 |
+
"grad_norm": 2.88493013381958,
|
| 974 |
+
"learning_rate": 2.5879247657670258e-05,
|
| 975 |
+
"loss": 4.8648,
|
| 976 |
+
"step": 13800
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.4859460215354496,
|
| 980 |
+
"grad_norm": 2.653207302093506,
|
| 981 |
+
"learning_rate": 2.570444693049923e-05,
|
| 982 |
+
"loss": 4.8199,
|
| 983 |
+
"step": 13900
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.4894420360788701,
|
| 987 |
+
"grad_norm": 2.7175774574279785,
|
| 988 |
+
"learning_rate": 2.5529646203328205e-05,
|
| 989 |
+
"loss": 4.8838,
|
| 990 |
+
"step": 14000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.49293805062229057,
|
| 994 |
+
"grad_norm": 2.9639148712158203,
|
| 995 |
+
"learning_rate": 2.535484547615718e-05,
|
| 996 |
+
"loss": 4.881,
|
| 997 |
+
"step": 14100
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.4964340651657111,
|
| 1001 |
+
"grad_norm": 2.8595130443573,
|
| 1002 |
+
"learning_rate": 2.518004474898616e-05,
|
| 1003 |
+
"loss": 4.8871,
|
| 1004 |
+
"step": 14200
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.4999300797091316,
|
| 1008 |
+
"grad_norm": 2.4576876163482666,
|
| 1009 |
+
"learning_rate": 2.5005244021815133e-05,
|
| 1010 |
+
"loss": 4.8357,
|
| 1011 |
+
"step": 14300
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5034260942525521,
|
| 1015 |
+
"grad_norm": 2.7865824699401855,
|
| 1016 |
+
"learning_rate": 2.4830443294644107e-05,
|
| 1017 |
+
"loss": 4.8026,
|
| 1018 |
+
"step": 14400
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.5069221087959725,
|
| 1022 |
+
"grad_norm": 2.741359233856201,
|
| 1023 |
+
"learning_rate": 2.465564256747308e-05,
|
| 1024 |
+
"loss": 4.8478,
|
| 1025 |
+
"step": 14500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5104181233393931,
|
| 1029 |
+
"grad_norm": 2.35073184967041,
|
| 1030 |
+
"learning_rate": 2.4480841840302057e-05,
|
| 1031 |
+
"loss": 4.8495,
|
| 1032 |
+
"step": 14600
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5139141378828136,
|
| 1036 |
+
"grad_norm": 2.672268867492676,
|
| 1037 |
+
"learning_rate": 2.430604111313103e-05,
|
| 1038 |
+
"loss": 4.8815,
|
| 1039 |
+
"step": 14700
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5174101524262341,
|
| 1043 |
+
"grad_norm": 2.589656114578247,
|
| 1044 |
+
"learning_rate": 2.4131240385960008e-05,
|
| 1045 |
+
"loss": 4.8814,
|
| 1046 |
+
"step": 14800
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5209061669696546,
|
| 1050 |
+
"grad_norm": 3.2942378520965576,
|
| 1051 |
+
"learning_rate": 2.3956439658788982e-05,
|
| 1052 |
+
"loss": 4.8035,
|
| 1053 |
+
"step": 14900
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.524402181513075,
|
| 1057 |
+
"grad_norm": 2.683692216873169,
|
| 1058 |
+
"learning_rate": 2.3781638931617956e-05,
|
| 1059 |
+
"loss": 4.8741,
|
| 1060 |
+
"step": 15000
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5278981960564956,
|
| 1064 |
+
"grad_norm": 2.480876922607422,
|
| 1065 |
+
"learning_rate": 2.3606838204446933e-05,
|
| 1066 |
+
"loss": 4.8208,
|
| 1067 |
+
"step": 15100
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5313942105999161,
|
| 1071 |
+
"grad_norm": 3.073580503463745,
|
| 1072 |
+
"learning_rate": 2.3432037477275906e-05,
|
| 1073 |
+
"loss": 4.8279,
|
| 1074 |
+
"step": 15200
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5348902251433366,
|
| 1078 |
+
"grad_norm": 3.04691219329834,
|
| 1079 |
+
"learning_rate": 2.3257236750104883e-05,
|
| 1080 |
+
"loss": 4.8546,
|
| 1081 |
+
"step": 15300
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5383862396867571,
|
| 1085 |
+
"grad_norm": 3.1862494945526123,
|
| 1086 |
+
"learning_rate": 2.3082436022933857e-05,
|
| 1087 |
+
"loss": 4.8761,
|
| 1088 |
+
"step": 15400
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.5418822542301776,
|
| 1092 |
+
"grad_norm": 3.067617654800415,
|
| 1093 |
+
"learning_rate": 2.290763529576283e-05,
|
| 1094 |
+
"loss": 4.8721,
|
| 1095 |
+
"step": 15500
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.5453782687735981,
|
| 1099 |
+
"grad_norm": 3.071998119354248,
|
| 1100 |
+
"learning_rate": 2.2732834568591808e-05,
|
| 1101 |
+
"loss": 4.8404,
|
| 1102 |
+
"step": 15600
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.5488742833170186,
|
| 1106 |
+
"grad_norm": 3.0393383502960205,
|
| 1107 |
+
"learning_rate": 2.255803384142078e-05,
|
| 1108 |
+
"loss": 4.8632,
|
| 1109 |
+
"step": 15700
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.5523702978604391,
|
| 1113 |
+
"grad_norm": 2.7928342819213867,
|
| 1114 |
+
"learning_rate": 2.238323311424976e-05,
|
| 1115 |
+
"loss": 4.8156,
|
| 1116 |
+
"step": 15800
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.5558663124038596,
|
| 1120 |
+
"grad_norm": 2.616582155227661,
|
| 1121 |
+
"learning_rate": 2.2208432387078732e-05,
|
| 1122 |
+
"loss": 4.8612,
|
| 1123 |
+
"step": 15900
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.55936232694728,
|
| 1127 |
+
"grad_norm": 2.9017162322998047,
|
| 1128 |
+
"learning_rate": 2.2033631659907706e-05,
|
| 1129 |
+
"loss": 4.8416,
|
| 1130 |
+
"step": 16000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.5628583414907006,
|
| 1134 |
+
"grad_norm": 3.112460136413574,
|
| 1135 |
+
"learning_rate": 2.1858830932736683e-05,
|
| 1136 |
+
"loss": 4.8362,
|
| 1137 |
+
"step": 16100
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.5663543560341211,
|
| 1141 |
+
"grad_norm": 2.453512668609619,
|
| 1142 |
+
"learning_rate": 2.1684030205565657e-05,
|
| 1143 |
+
"loss": 4.8558,
|
| 1144 |
+
"step": 16200
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.5698503705775416,
|
| 1148 |
+
"grad_norm": 2.6998403072357178,
|
| 1149 |
+
"learning_rate": 2.1509229478394634e-05,
|
| 1150 |
+
"loss": 4.8444,
|
| 1151 |
+
"step": 16300
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.5733463851209621,
|
| 1155 |
+
"grad_norm": 3.0294365882873535,
|
| 1156 |
+
"learning_rate": 2.1334428751223607e-05,
|
| 1157 |
+
"loss": 4.8883,
|
| 1158 |
+
"step": 16400
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.5768423996643826,
|
| 1162 |
+
"grad_norm": 3.008953809738159,
|
| 1163 |
+
"learning_rate": 2.115962802405258e-05,
|
| 1164 |
+
"loss": 4.8333,
|
| 1165 |
+
"step": 16500
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.5803384142078031,
|
| 1169 |
+
"grad_norm": 3.1423451900482178,
|
| 1170 |
+
"learning_rate": 2.0984827296881558e-05,
|
| 1171 |
+
"loss": 4.8312,
|
| 1172 |
+
"step": 16600
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.5838344287512236,
|
| 1176 |
+
"grad_norm": 2.988421678543091,
|
| 1177 |
+
"learning_rate": 2.0810026569710532e-05,
|
| 1178 |
+
"loss": 4.8608,
|
| 1179 |
+
"step": 16700
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.5873304432946441,
|
| 1183 |
+
"grad_norm": 2.9294745922088623,
|
| 1184 |
+
"learning_rate": 2.063522584253951e-05,
|
| 1185 |
+
"loss": 4.8528,
|
| 1186 |
+
"step": 16800
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.5908264578380646,
|
| 1190 |
+
"grad_norm": 2.831146478652954,
|
| 1191 |
+
"learning_rate": 2.0460425115368483e-05,
|
| 1192 |
+
"loss": 4.8472,
|
| 1193 |
+
"step": 16900
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.5943224723814851,
|
| 1197 |
+
"grad_norm": 2.5871362686157227,
|
| 1198 |
+
"learning_rate": 2.0285624388197456e-05,
|
| 1199 |
+
"loss": 4.8068,
|
| 1200 |
+
"step": 17000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.5978184869249056,
|
| 1204 |
+
"grad_norm": 2.6208789348602295,
|
| 1205 |
+
"learning_rate": 2.0110823661026433e-05,
|
| 1206 |
+
"loss": 4.8573,
|
| 1207 |
+
"step": 17100
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.6013145014683261,
|
| 1211 |
+
"grad_norm": 3.1844122409820557,
|
| 1212 |
+
"learning_rate": 1.9936022933855407e-05,
|
| 1213 |
+
"loss": 4.8837,
|
| 1214 |
+
"step": 17200
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6048105160117466,
|
| 1218 |
+
"grad_norm": 2.6045706272125244,
|
| 1219 |
+
"learning_rate": 1.976122220668438e-05,
|
| 1220 |
+
"loss": 4.8873,
|
| 1221 |
+
"step": 17300
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6083065305551671,
|
| 1225 |
+
"grad_norm": 2.6569807529449463,
|
| 1226 |
+
"learning_rate": 1.9586421479513354e-05,
|
| 1227 |
+
"loss": 4.8533,
|
| 1228 |
+
"step": 17400
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6118025450985876,
|
| 1232 |
+
"grad_norm": 2.7514195442199707,
|
| 1233 |
+
"learning_rate": 1.941162075234233e-05,
|
| 1234 |
+
"loss": 4.8454,
|
| 1235 |
+
"step": 17500
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6152985596420081,
|
| 1239 |
+
"grad_norm": 3.129321575164795,
|
| 1240 |
+
"learning_rate": 1.9236820025171305e-05,
|
| 1241 |
+
"loss": 4.8622,
|
| 1242 |
+
"step": 17600
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6187945741854286,
|
| 1246 |
+
"grad_norm": 2.610549211502075,
|
| 1247 |
+
"learning_rate": 1.906201929800028e-05,
|
| 1248 |
+
"loss": 4.7976,
|
| 1249 |
+
"step": 17700
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6222905887288491,
|
| 1253 |
+
"grad_norm": 2.735844135284424,
|
| 1254 |
+
"learning_rate": 1.8887218570829256e-05,
|
| 1255 |
+
"loss": 4.8136,
|
| 1256 |
+
"step": 17800
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6257866032722696,
|
| 1260 |
+
"grad_norm": 3.020894765853882,
|
| 1261 |
+
"learning_rate": 1.871241784365823e-05,
|
| 1262 |
+
"loss": 4.8348,
|
| 1263 |
+
"step": 17900
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6292826178156901,
|
| 1267 |
+
"grad_norm": 3.2946112155914307,
|
| 1268 |
+
"learning_rate": 1.8537617116487203e-05,
|
| 1269 |
+
"loss": 4.8435,
|
| 1270 |
+
"step": 18000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6327786323591106,
|
| 1274 |
+
"grad_norm": 2.6459012031555176,
|
| 1275 |
+
"learning_rate": 1.836281638931618e-05,
|
| 1276 |
+
"loss": 4.8705,
|
| 1277 |
+
"step": 18100
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.6362746469025311,
|
| 1281 |
+
"grad_norm": 3.0113513469696045,
|
| 1282 |
+
"learning_rate": 1.8188015662145154e-05,
|
| 1283 |
+
"loss": 4.8347,
|
| 1284 |
+
"step": 18200
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.6397706614459516,
|
| 1288 |
+
"grad_norm": 2.8123104572296143,
|
| 1289 |
+
"learning_rate": 1.8013214934974128e-05,
|
| 1290 |
+
"loss": 4.8568,
|
| 1291 |
+
"step": 18300
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.6432666759893721,
|
| 1295 |
+
"grad_norm": 2.8283095359802246,
|
| 1296 |
+
"learning_rate": 1.7838414207803105e-05,
|
| 1297 |
+
"loss": 4.8248,
|
| 1298 |
+
"step": 18400
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.6467626905327926,
|
| 1302 |
+
"grad_norm": 3.098484516143799,
|
| 1303 |
+
"learning_rate": 1.766361348063208e-05,
|
| 1304 |
+
"loss": 4.825,
|
| 1305 |
+
"step": 18500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.6502587050762131,
|
| 1309 |
+
"grad_norm": 2.566540002822876,
|
| 1310 |
+
"learning_rate": 1.7488812753461055e-05,
|
| 1311 |
+
"loss": 4.7884,
|
| 1312 |
+
"step": 18600
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.6537547196196336,
|
| 1316 |
+
"grad_norm": 2.4459304809570312,
|
| 1317 |
+
"learning_rate": 1.731401202629003e-05,
|
| 1318 |
+
"loss": 4.8451,
|
| 1319 |
+
"step": 18700
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.6572507341630541,
|
| 1323 |
+
"grad_norm": 3.134326457977295,
|
| 1324 |
+
"learning_rate": 1.7139211299119003e-05,
|
| 1325 |
+
"loss": 4.8586,
|
| 1326 |
+
"step": 18800
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.6607467487064747,
|
| 1330 |
+
"grad_norm": 2.64436674118042,
|
| 1331 |
+
"learning_rate": 1.696441057194798e-05,
|
| 1332 |
+
"loss": 4.8463,
|
| 1333 |
+
"step": 18900
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.6642427632498952,
|
| 1337 |
+
"grad_norm": 2.925896167755127,
|
| 1338 |
+
"learning_rate": 1.6789609844776954e-05,
|
| 1339 |
+
"loss": 4.8216,
|
| 1340 |
+
"step": 19000
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.6677387777933156,
|
| 1344 |
+
"grad_norm": 2.873636245727539,
|
| 1345 |
+
"learning_rate": 1.661480911760593e-05,
|
| 1346 |
+
"loss": 4.8579,
|
| 1347 |
+
"step": 19100
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.6712347923367361,
|
| 1351 |
+
"grad_norm": 2.8140294551849365,
|
| 1352 |
+
"learning_rate": 1.6440008390434904e-05,
|
| 1353 |
+
"loss": 4.8282,
|
| 1354 |
+
"step": 19200
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.6747308068801566,
|
| 1358 |
+
"grad_norm": 2.66567325592041,
|
| 1359 |
+
"learning_rate": 1.6265207663263878e-05,
|
| 1360 |
+
"loss": 4.8042,
|
| 1361 |
+
"step": 19300
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.6782268214235772,
|
| 1365 |
+
"grad_norm": 2.895731210708618,
|
| 1366 |
+
"learning_rate": 1.6090406936092855e-05,
|
| 1367 |
+
"loss": 4.81,
|
| 1368 |
+
"step": 19400
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.6817228359669977,
|
| 1372 |
+
"grad_norm": 2.673616886138916,
|
| 1373 |
+
"learning_rate": 1.591560620892183e-05,
|
| 1374 |
+
"loss": 4.8267,
|
| 1375 |
+
"step": 19500
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.6852188505104181,
|
| 1379 |
+
"grad_norm": 2.8910722732543945,
|
| 1380 |
+
"learning_rate": 1.5740805481750806e-05,
|
| 1381 |
+
"loss": 4.8449,
|
| 1382 |
+
"step": 19600
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.6887148650538386,
|
| 1386 |
+
"grad_norm": 3.0458858013153076,
|
| 1387 |
+
"learning_rate": 1.556600475457978e-05,
|
| 1388 |
+
"loss": 4.8022,
|
| 1389 |
+
"step": 19700
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.6922108795972591,
|
| 1393 |
+
"grad_norm": 2.440716028213501,
|
| 1394 |
+
"learning_rate": 1.5391204027408753e-05,
|
| 1395 |
+
"loss": 4.8125,
|
| 1396 |
+
"step": 19800
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.6957068941406797,
|
| 1400 |
+
"grad_norm": 2.811601161956787,
|
| 1401 |
+
"learning_rate": 1.521640330023773e-05,
|
| 1402 |
+
"loss": 4.8567,
|
| 1403 |
+
"step": 19900
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.6992029086841002,
|
| 1407 |
+
"grad_norm": 2.6201772689819336,
|
| 1408 |
+
"learning_rate": 1.5041602573066704e-05,
|
| 1409 |
+
"loss": 4.8375,
|
| 1410 |
+
"step": 20000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7026989232275206,
|
| 1414 |
+
"grad_norm": 3.1684257984161377,
|
| 1415 |
+
"learning_rate": 1.4866801845895681e-05,
|
| 1416 |
+
"loss": 4.8707,
|
| 1417 |
+
"step": 20100
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7061949377709411,
|
| 1421 |
+
"grad_norm": 2.945105791091919,
|
| 1422 |
+
"learning_rate": 1.4692001118724655e-05,
|
| 1423 |
+
"loss": 4.8065,
|
| 1424 |
+
"step": 20200
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7096909523143616,
|
| 1428 |
+
"grad_norm": 2.767002820968628,
|
| 1429 |
+
"learning_rate": 1.4517200391553628e-05,
|
| 1430 |
+
"loss": 4.8594,
|
| 1431 |
+
"step": 20300
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7131869668577822,
|
| 1435 |
+
"grad_norm": 2.928342819213867,
|
| 1436 |
+
"learning_rate": 1.4342399664382605e-05,
|
| 1437 |
+
"loss": 4.8418,
|
| 1438 |
+
"step": 20400
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7166829814012027,
|
| 1442 |
+
"grad_norm": 2.7073936462402344,
|
| 1443 |
+
"learning_rate": 1.4167598937211579e-05,
|
| 1444 |
+
"loss": 4.8515,
|
| 1445 |
+
"step": 20500
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7201789959446231,
|
| 1449 |
+
"grad_norm": 3.105440616607666,
|
| 1450 |
+
"learning_rate": 1.3992798210040556e-05,
|
| 1451 |
+
"loss": 4.8075,
|
| 1452 |
+
"step": 20600
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7236750104880436,
|
| 1456 |
+
"grad_norm": 2.8033523559570312,
|
| 1457 |
+
"learning_rate": 1.381799748286953e-05,
|
| 1458 |
+
"loss": 4.8108,
|
| 1459 |
+
"step": 20700
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.7271710250314641,
|
| 1463 |
+
"grad_norm": 2.9374163150787354,
|
| 1464 |
+
"learning_rate": 1.3643196755698503e-05,
|
| 1465 |
+
"loss": 4.8197,
|
| 1466 |
+
"step": 20800
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.7306670395748847,
|
| 1470 |
+
"grad_norm": 2.5617268085479736,
|
| 1471 |
+
"learning_rate": 1.346839602852748e-05,
|
| 1472 |
+
"loss": 4.8105,
|
| 1473 |
+
"step": 20900
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.7341630541183052,
|
| 1477 |
+
"grad_norm": 2.8728411197662354,
|
| 1478 |
+
"learning_rate": 1.3293595301356454e-05,
|
| 1479 |
+
"loss": 4.8447,
|
| 1480 |
+
"step": 21000
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.7376590686617256,
|
| 1484 |
+
"grad_norm": 2.78499436378479,
|
| 1485 |
+
"learning_rate": 1.3118794574185428e-05,
|
| 1486 |
+
"loss": 4.79,
|
| 1487 |
+
"step": 21100
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.7411550832051461,
|
| 1491 |
+
"grad_norm": 2.84222412109375,
|
| 1492 |
+
"learning_rate": 1.2943993847014405e-05,
|
| 1493 |
+
"loss": 4.8189,
|
| 1494 |
+
"step": 21200
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.7446510977485666,
|
| 1498 |
+
"grad_norm": 2.6608426570892334,
|
| 1499 |
+
"learning_rate": 1.2769193119843379e-05,
|
| 1500 |
+
"loss": 4.7787,
|
| 1501 |
+
"step": 21300
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.7481471122919872,
|
| 1505 |
+
"grad_norm": 2.662895917892456,
|
| 1506 |
+
"learning_rate": 1.2594392392672356e-05,
|
| 1507 |
+
"loss": 4.8287,
|
| 1508 |
+
"step": 21400
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.7516431268354077,
|
| 1512 |
+
"grad_norm": 2.7515127658843994,
|
| 1513 |
+
"learning_rate": 1.241959166550133e-05,
|
| 1514 |
+
"loss": 4.7988,
|
| 1515 |
+
"step": 21500
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.7551391413788281,
|
| 1519 |
+
"grad_norm": 2.9696640968322754,
|
| 1520 |
+
"learning_rate": 1.2244790938330305e-05,
|
| 1521 |
+
"loss": 4.82,
|
| 1522 |
+
"step": 21600
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.7586351559222486,
|
| 1526 |
+
"grad_norm": 3.002441644668579,
|
| 1527 |
+
"learning_rate": 1.206999021115928e-05,
|
| 1528 |
+
"loss": 4.7711,
|
| 1529 |
+
"step": 21700
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.7621311704656691,
|
| 1533 |
+
"grad_norm": 3.171853542327881,
|
| 1534 |
+
"learning_rate": 1.1895189483988254e-05,
|
| 1535 |
+
"loss": 4.8068,
|
| 1536 |
+
"step": 21800
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.7656271850090897,
|
| 1540 |
+
"grad_norm": 2.604813575744629,
|
| 1541 |
+
"learning_rate": 1.172038875681723e-05,
|
| 1542 |
+
"loss": 4.842,
|
| 1543 |
+
"step": 21900
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.7691231995525102,
|
| 1547 |
+
"grad_norm": 3.086451768875122,
|
| 1548 |
+
"learning_rate": 1.1545588029646205e-05,
|
| 1549 |
+
"loss": 4.8064,
|
| 1550 |
+
"step": 22000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.7726192140959306,
|
| 1554 |
+
"grad_norm": 3.1282565593719482,
|
| 1555 |
+
"learning_rate": 1.1370787302475178e-05,
|
| 1556 |
+
"loss": 4.827,
|
| 1557 |
+
"step": 22100
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.7761152286393511,
|
| 1561 |
+
"grad_norm": 2.6735763549804688,
|
| 1562 |
+
"learning_rate": 1.1195986575304154e-05,
|
| 1563 |
+
"loss": 4.792,
|
| 1564 |
+
"step": 22200
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.7796112431827716,
|
| 1568 |
+
"grad_norm": 2.88252329826355,
|
| 1569 |
+
"learning_rate": 1.1021185848133129e-05,
|
| 1570 |
+
"loss": 4.8025,
|
| 1571 |
+
"step": 22300
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.7831072577261922,
|
| 1575 |
+
"grad_norm": 3.1057016849517822,
|
| 1576 |
+
"learning_rate": 1.0846385120962103e-05,
|
| 1577 |
+
"loss": 4.8332,
|
| 1578 |
+
"step": 22400
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.7866032722696127,
|
| 1582 |
+
"grad_norm": 2.930310010910034,
|
| 1583 |
+
"learning_rate": 1.0671584393791078e-05,
|
| 1584 |
+
"loss": 4.8156,
|
| 1585 |
+
"step": 22500
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.7900992868130331,
|
| 1589 |
+
"grad_norm": 2.785041093826294,
|
| 1590 |
+
"learning_rate": 1.0496783666620053e-05,
|
| 1591 |
+
"loss": 4.8164,
|
| 1592 |
+
"step": 22600
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.7935953013564536,
|
| 1596 |
+
"grad_norm": 3.089155912399292,
|
| 1597 |
+
"learning_rate": 1.0321982939449029e-05,
|
| 1598 |
+
"loss": 4.8126,
|
| 1599 |
+
"step": 22700
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.7970913158998741,
|
| 1603 |
+
"grad_norm": 2.8560636043548584,
|
| 1604 |
+
"learning_rate": 1.0147182212278002e-05,
|
| 1605 |
+
"loss": 4.8162,
|
| 1606 |
+
"step": 22800
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8005873304432947,
|
| 1610 |
+
"grad_norm": 2.9523401260375977,
|
| 1611 |
+
"learning_rate": 9.972381485106978e-06,
|
| 1612 |
+
"loss": 4.8756,
|
| 1613 |
+
"step": 22900
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8040833449867152,
|
| 1617 |
+
"grad_norm": 2.6021039485931396,
|
| 1618 |
+
"learning_rate": 9.797580757935953e-06,
|
| 1619 |
+
"loss": 4.8416,
|
| 1620 |
+
"step": 23000
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8075793595301356,
|
| 1624 |
+
"grad_norm": 2.580181121826172,
|
| 1625 |
+
"learning_rate": 9.622780030764929e-06,
|
| 1626 |
+
"loss": 4.8208,
|
| 1627 |
+
"step": 23100
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8110753740735561,
|
| 1631 |
+
"grad_norm": 2.7473840713500977,
|
| 1632 |
+
"learning_rate": 9.447979303593904e-06,
|
| 1633 |
+
"loss": 4.8074,
|
| 1634 |
+
"step": 23200
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8145713886169766,
|
| 1638 |
+
"grad_norm": 2.52461314201355,
|
| 1639 |
+
"learning_rate": 9.273178576422878e-06,
|
| 1640 |
+
"loss": 4.8019,
|
| 1641 |
+
"step": 23300
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.8180674031603972,
|
| 1645 |
+
"grad_norm": 3.13628888130188,
|
| 1646 |
+
"learning_rate": 9.098377849251853e-06,
|
| 1647 |
+
"loss": 4.8007,
|
| 1648 |
+
"step": 23400
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.8215634177038177,
|
| 1652 |
+
"grad_norm": 2.702481746673584,
|
| 1653 |
+
"learning_rate": 8.923577122080828e-06,
|
| 1654 |
+
"loss": 4.8039,
|
| 1655 |
+
"step": 23500
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.8250594322472381,
|
| 1659 |
+
"grad_norm": 3.0160887241363525,
|
| 1660 |
+
"learning_rate": 8.748776394909804e-06,
|
| 1661 |
+
"loss": 4.8218,
|
| 1662 |
+
"step": 23600
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.8285554467906586,
|
| 1666 |
+
"grad_norm": 3.1599791049957275,
|
| 1667 |
+
"learning_rate": 8.573975667738777e-06,
|
| 1668 |
+
"loss": 4.7874,
|
| 1669 |
+
"step": 23700
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.8320514613340791,
|
| 1673 |
+
"grad_norm": 2.9699933528900146,
|
| 1674 |
+
"learning_rate": 8.399174940567753e-06,
|
| 1675 |
+
"loss": 4.8316,
|
| 1676 |
+
"step": 23800
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.8355474758774997,
|
| 1680 |
+
"grad_norm": 2.7698373794555664,
|
| 1681 |
+
"learning_rate": 8.224374213396728e-06,
|
| 1682 |
+
"loss": 4.838,
|
| 1683 |
+
"step": 23900
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.8390434904209202,
|
| 1687 |
+
"grad_norm": 2.6946427822113037,
|
| 1688 |
+
"learning_rate": 8.049573486225704e-06,
|
| 1689 |
+
"loss": 4.7893,
|
| 1690 |
+
"step": 24000
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.8425395049643406,
|
| 1694 |
+
"grad_norm": 3.0160024166107178,
|
| 1695 |
+
"learning_rate": 7.874772759054679e-06,
|
| 1696 |
+
"loss": 4.8058,
|
| 1697 |
+
"step": 24100
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.8460355195077611,
|
| 1701 |
+
"grad_norm": 2.491325855255127,
|
| 1702 |
+
"learning_rate": 7.699972031883653e-06,
|
| 1703 |
+
"loss": 4.7965,
|
| 1704 |
+
"step": 24200
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.8495315340511816,
|
| 1708 |
+
"grad_norm": 3.0036044120788574,
|
| 1709 |
+
"learning_rate": 7.525171304712628e-06,
|
| 1710 |
+
"loss": 4.7869,
|
| 1711 |
+
"step": 24300
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.8530275485946022,
|
| 1715 |
+
"grad_norm": 3.335988759994507,
|
| 1716 |
+
"learning_rate": 7.350370577541603e-06,
|
| 1717 |
+
"loss": 4.8458,
|
| 1718 |
+
"step": 24400
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.8565235631380227,
|
| 1722 |
+
"grad_norm": 2.7462995052337646,
|
| 1723 |
+
"learning_rate": 7.175569850370578e-06,
|
| 1724 |
+
"loss": 4.7888,
|
| 1725 |
+
"step": 24500
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.8600195776814431,
|
| 1729 |
+
"grad_norm": 2.402048110961914,
|
| 1730 |
+
"learning_rate": 7.000769123199553e-06,
|
| 1731 |
+
"loss": 4.7661,
|
| 1732 |
+
"step": 24600
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.8635155922248636,
|
| 1736 |
+
"grad_norm": 2.752999782562256,
|
| 1737 |
+
"learning_rate": 6.825968396028527e-06,
|
| 1738 |
+
"loss": 4.8055,
|
| 1739 |
+
"step": 24700
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.8670116067682841,
|
| 1743 |
+
"grad_norm": 2.9346210956573486,
|
| 1744 |
+
"learning_rate": 6.651167668857502e-06,
|
| 1745 |
+
"loss": 4.7676,
|
| 1746 |
+
"step": 24800
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.8705076213117047,
|
| 1750 |
+
"grad_norm": 2.790853261947632,
|
| 1751 |
+
"learning_rate": 6.476366941686478e-06,
|
| 1752 |
+
"loss": 4.7821,
|
| 1753 |
+
"step": 24900
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.8740036358551252,
|
| 1757 |
+
"grad_norm": 2.6606948375701904,
|
| 1758 |
+
"learning_rate": 6.301566214515453e-06,
|
| 1759 |
+
"loss": 4.7845,
|
| 1760 |
+
"step": 25000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.8774996503985456,
|
| 1764 |
+
"grad_norm": 2.6995749473571777,
|
| 1765 |
+
"learning_rate": 6.1267654873444275e-06,
|
| 1766 |
+
"loss": 4.8264,
|
| 1767 |
+
"step": 25100
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.8809956649419661,
|
| 1771 |
+
"grad_norm": 2.790954351425171,
|
| 1772 |
+
"learning_rate": 5.951964760173403e-06,
|
| 1773 |
+
"loss": 4.8493,
|
| 1774 |
+
"step": 25200
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.8844916794853867,
|
| 1778 |
+
"grad_norm": 2.903285264968872,
|
| 1779 |
+
"learning_rate": 5.777164033002377e-06,
|
| 1780 |
+
"loss": 4.8131,
|
| 1781 |
+
"step": 25300
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.8879876940288072,
|
| 1785 |
+
"grad_norm": 2.517873525619507,
|
| 1786 |
+
"learning_rate": 5.602363305831353e-06,
|
| 1787 |
+
"loss": 4.7983,
|
| 1788 |
+
"step": 25400
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.8914837085722277,
|
| 1792 |
+
"grad_norm": 2.7883822917938232,
|
| 1793 |
+
"learning_rate": 5.427562578660327e-06,
|
| 1794 |
+
"loss": 4.778,
|
| 1795 |
+
"step": 25500
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.8949797231156481,
|
| 1799 |
+
"grad_norm": 3.072084426879883,
|
| 1800 |
+
"learning_rate": 5.252761851489303e-06,
|
| 1801 |
+
"loss": 4.8017,
|
| 1802 |
+
"step": 25600
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.8984757376590686,
|
| 1806 |
+
"grad_norm": 2.7740066051483154,
|
| 1807 |
+
"learning_rate": 5.077961124318278e-06,
|
| 1808 |
+
"loss": 4.8314,
|
| 1809 |
+
"step": 25700
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9019717522024892,
|
| 1813 |
+
"grad_norm": 2.706472873687744,
|
| 1814 |
+
"learning_rate": 4.903160397147253e-06,
|
| 1815 |
+
"loss": 4.7765,
|
| 1816 |
+
"step": 25800
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9054677667459097,
|
| 1820 |
+
"grad_norm": 3.1904239654541016,
|
| 1821 |
+
"learning_rate": 4.728359669976228e-06,
|
| 1822 |
+
"loss": 4.8255,
|
| 1823 |
+
"step": 25900
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9089637812893302,
|
| 1827 |
+
"grad_norm": 2.787018299102783,
|
| 1828 |
+
"learning_rate": 4.5535589428052025e-06,
|
| 1829 |
+
"loss": 4.8399,
|
| 1830 |
+
"step": 26000
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.9124597958327507,
|
| 1834 |
+
"grad_norm": 2.6253585815429688,
|
| 1835 |
+
"learning_rate": 4.378758215634177e-06,
|
| 1836 |
+
"loss": 4.8217,
|
| 1837 |
+
"step": 26100
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.9159558103761711,
|
| 1841 |
+
"grad_norm": 2.7230615615844727,
|
| 1842 |
+
"learning_rate": 4.2039574884631515e-06,
|
| 1843 |
+
"loss": 4.7865,
|
| 1844 |
+
"step": 26200
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.9194518249195917,
|
| 1848 |
+
"grad_norm": 3.5472490787506104,
|
| 1849 |
+
"learning_rate": 4.029156761292127e-06,
|
| 1850 |
+
"loss": 4.7454,
|
| 1851 |
+
"step": 26300
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.9229478394630122,
|
| 1855 |
+
"grad_norm": 2.610646963119507,
|
| 1856 |
+
"learning_rate": 3.854356034121102e-06,
|
| 1857 |
+
"loss": 4.7943,
|
| 1858 |
+
"step": 26400
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.9264438540064327,
|
| 1862 |
+
"grad_norm": 2.580254316329956,
|
| 1863 |
+
"learning_rate": 3.679555306950077e-06,
|
| 1864 |
+
"loss": 4.8029,
|
| 1865 |
+
"step": 26500
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.9299398685498532,
|
| 1869 |
+
"grad_norm": 2.8292527198791504,
|
| 1870 |
+
"learning_rate": 3.504754579779052e-06,
|
| 1871 |
+
"loss": 4.8067,
|
| 1872 |
+
"step": 26600
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.9334358830932736,
|
| 1876 |
+
"grad_norm": 2.95322322845459,
|
| 1877 |
+
"learning_rate": 3.3299538526080267e-06,
|
| 1878 |
+
"loss": 4.7623,
|
| 1879 |
+
"step": 26700
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.9369318976366942,
|
| 1883 |
+
"grad_norm": 2.604645013809204,
|
| 1884 |
+
"learning_rate": 3.155153125437002e-06,
|
| 1885 |
+
"loss": 4.8072,
|
| 1886 |
+
"step": 26800
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.9404279121801147,
|
| 1890 |
+
"grad_norm": 2.9846179485321045,
|
| 1891 |
+
"learning_rate": 2.980352398265977e-06,
|
| 1892 |
+
"loss": 4.7879,
|
| 1893 |
+
"step": 26900
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.9439239267235352,
|
| 1897 |
+
"grad_norm": 2.7782154083251953,
|
| 1898 |
+
"learning_rate": 2.805551671094952e-06,
|
| 1899 |
+
"loss": 4.7872,
|
| 1900 |
+
"step": 27000
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.9474199412669557,
|
| 1904 |
+
"grad_norm": 3.231360673904419,
|
| 1905 |
+
"learning_rate": 2.630750943923927e-06,
|
| 1906 |
+
"loss": 4.8513,
|
| 1907 |
+
"step": 27100
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.9509159558103761,
|
| 1911 |
+
"grad_norm": 2.78902006149292,
|
| 1912 |
+
"learning_rate": 2.455950216752902e-06,
|
| 1913 |
+
"loss": 4.8163,
|
| 1914 |
+
"step": 27200
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.9544119703537967,
|
| 1918 |
+
"grad_norm": 2.885953903198242,
|
| 1919 |
+
"learning_rate": 2.281149489581877e-06,
|
| 1920 |
+
"loss": 4.7673,
|
| 1921 |
+
"step": 27300
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.9579079848972172,
|
| 1925 |
+
"grad_norm": 3.1494991779327393,
|
| 1926 |
+
"learning_rate": 2.1063487624108518e-06,
|
| 1927 |
+
"loss": 4.7808,
|
| 1928 |
+
"step": 27400
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.9614039994406377,
|
| 1932 |
+
"grad_norm": 2.541231870651245,
|
| 1933 |
+
"learning_rate": 1.9315480352398267e-06,
|
| 1934 |
+
"loss": 4.8166,
|
| 1935 |
+
"step": 27500
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.9649000139840582,
|
| 1939 |
+
"grad_norm": 2.773287773132324,
|
| 1940 |
+
"learning_rate": 1.7567473080688017e-06,
|
| 1941 |
+
"loss": 4.7576,
|
| 1942 |
+
"step": 27600
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.9683960285274786,
|
| 1946 |
+
"grad_norm": 2.559238910675049,
|
| 1947 |
+
"learning_rate": 1.5819465808977766e-06,
|
| 1948 |
+
"loss": 4.7819,
|
| 1949 |
+
"step": 27700
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.9718920430708992,
|
| 1953 |
+
"grad_norm": 2.684382200241089,
|
| 1954 |
+
"learning_rate": 1.4071458537267516e-06,
|
| 1955 |
+
"loss": 4.783,
|
| 1956 |
+
"step": 27800
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.9753880576143197,
|
| 1960 |
+
"grad_norm": 2.598681688308716,
|
| 1961 |
+
"learning_rate": 1.2323451265557265e-06,
|
| 1962 |
+
"loss": 4.785,
|
| 1963 |
+
"step": 27900
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.9788840721577402,
|
| 1967 |
+
"grad_norm": 2.9225780963897705,
|
| 1968 |
+
"learning_rate": 1.0575443993847015e-06,
|
| 1969 |
+
"loss": 4.7619,
|
| 1970 |
+
"step": 28000
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.9823800867011607,
|
| 1974 |
+
"grad_norm": 2.758411407470703,
|
| 1975 |
+
"learning_rate": 8.827436722136765e-07,
|
| 1976 |
+
"loss": 4.8523,
|
| 1977 |
+
"step": 28100
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.9858761012445811,
|
| 1981 |
+
"grad_norm": 3.0110673904418945,
|
| 1982 |
+
"learning_rate": 7.079429450426515e-07,
|
| 1983 |
+
"loss": 4.7437,
|
| 1984 |
+
"step": 28200
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.9893721157880017,
|
| 1988 |
+
"grad_norm": 2.6037356853485107,
|
| 1989 |
+
"learning_rate": 5.331422178716264e-07,
|
| 1990 |
+
"loss": 4.7638,
|
| 1991 |
+
"step": 28300
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.9928681303314222,
|
| 1995 |
+
"grad_norm": 3.267136335372925,
|
| 1996 |
+
"learning_rate": 3.5834149070060136e-07,
|
| 1997 |
+
"loss": 4.7422,
|
| 1998 |
+
"step": 28400
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.9963641448748427,
|
| 2002 |
+
"grad_norm": 2.53847599029541,
|
| 2003 |
+
"learning_rate": 1.835407635295763e-07,
|
| 2004 |
+
"loss": 4.8052,
|
| 2005 |
+
"step": 28500
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.9998601594182632,
|
| 2009 |
+
"grad_norm": 2.709245204925537,
|
| 2010 |
+
"learning_rate": 8.740036358551251e-09,
|
| 2011 |
+
"loss": 4.7659,
|
| 2012 |
+
"step": 28600
|
| 2013 |
+
}
|
| 2014 |
+
],
|
| 2015 |
+
"logging_steps": 100,
|
| 2016 |
+
"max_steps": 28604,
|
| 2017 |
+
"num_input_tokens_seen": 0,
|
| 2018 |
+
"num_train_epochs": 1,
|
| 2019 |
+
"save_steps": 500,
|
| 2020 |
+
"stateful_callbacks": {
|
| 2021 |
+
"TrainerControl": {
|
| 2022 |
+
"args": {
|
| 2023 |
+
"should_epoch_stop": false,
|
| 2024 |
+
"should_evaluate": false,
|
| 2025 |
+
"should_log": false,
|
| 2026 |
+
"should_save": true,
|
| 2027 |
+
"should_training_stop": true
|
| 2028 |
+
},
|
| 2029 |
+
"attributes": {}
|
| 2030 |
+
}
|
| 2031 |
+
},
|
| 2032 |
+
"total_flos": 4.2974831149056e+16,
|
| 2033 |
+
"train_batch_size": 23,
|
| 2034 |
+
"trial_name": null,
|
| 2035 |
+
"trial_params": null
|
| 2036 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5219e41e4dceacec8a12f41d37a63963f260daaa24f2ceec3aa4813f0f19a9b
|
| 3 |
+
size 5368
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|