Upload model finetuned on codet5p-220m using strategy src_fm_fc_dctx
Browse files- README.md +50 -3
- added_tokens.json +8 -0
- config.json +33 -0
- generation_config.json +8 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +59 -0
- tokenizer.json +0 -0
- tokenizer_config.json +113 -0
- trainer_state.json +1603 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
CHANGED
@@ -1,3 +1,50 @@
|
|
1 |
-
---
|
2 |
-
license:
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: bsd-3-clause
|
3 |
+
base_model: Salesforce/codet5p-220m
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
model-index:
|
7 |
+
- name: finetuning_01_codet5p_src_fm_fc_dctx
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
+
should probably proofread and complete it, then remove this comment. -->
|
13 |
+
|
14 |
+
# finetuning_01_codet5p_src_fm_fc_dctx
|
15 |
+
|
16 |
+
This model is a fine-tuned version of [Salesforce/codet5p-220m](https://huggingface.co/Salesforce/codet5p-220m) on an unknown dataset.
|
17 |
+
|
18 |
+
## Model description
|
19 |
+
|
20 |
+
More information needed
|
21 |
+
|
22 |
+
## Intended uses & limitations
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Training and evaluation data
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training procedure
|
31 |
+
|
32 |
+
### Training hyperparameters
|
33 |
+
|
34 |
+
The following hyperparameters were used during training:
|
35 |
+
- learning_rate: 5e-05
|
36 |
+
- train_batch_size: 32
|
37 |
+
- eval_batch_size: 64
|
38 |
+
- seed: 42
|
39 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
40 |
+
- lr_scheduler_type: linear
|
41 |
+
- lr_scheduler_warmup_steps: 800
|
42 |
+
- num_epochs: 5
|
43 |
+
- mixed_precision_training: Native AMP
|
44 |
+
|
45 |
+
### Framework versions
|
46 |
+
|
47 |
+
- Transformers 4.38.2
|
48 |
+
- Pytorch 2.1.0
|
49 |
+
- Datasets 2.20.0
|
50 |
+
- Tokenizers 0.15.2
|
added_tokens.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"</ECTX>": 32103,
|
3 |
+
"</FCTX>": 32101,
|
4 |
+
"</PRIVATE_FCTX>": 32105,
|
5 |
+
"<ECTX>": 32102,
|
6 |
+
"<FCTX>": 32100,
|
7 |
+
"<PRIVATE_FCTX>": 32104
|
8 |
+
}
|
config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Salesforce/codet5p-220m",
|
3 |
+
"architectures": [
|
4 |
+
"T5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"bos_token_id": 1,
|
7 |
+
"classifier_dropout": 0.0,
|
8 |
+
"d_ff": 3072,
|
9 |
+
"d_kv": 64,
|
10 |
+
"d_model": 768,
|
11 |
+
"decoder_start_token_id": 0,
|
12 |
+
"dense_act_fn": "relu",
|
13 |
+
"dropout_rate": 0.1,
|
14 |
+
"eos_token_id": 2,
|
15 |
+
"feed_forward_proj": "relu",
|
16 |
+
"initializer_factor": 1.0,
|
17 |
+
"is_encoder_decoder": true,
|
18 |
+
"is_gated_act": false,
|
19 |
+
"layer_norm_epsilon": 1e-06,
|
20 |
+
"model_type": "t5",
|
21 |
+
"n_positions": 512,
|
22 |
+
"num_decoder_layers": 12,
|
23 |
+
"num_heads": 12,
|
24 |
+
"num_layers": 12,
|
25 |
+
"output_past": true,
|
26 |
+
"pad_token_id": 0,
|
27 |
+
"relative_attention_max_distance": 128,
|
28 |
+
"relative_attention_num_buckets": 32,
|
29 |
+
"torch_dtype": "float32",
|
30 |
+
"transformers_version": "4.38.2",
|
31 |
+
"use_cache": true,
|
32 |
+
"vocab_size": 32106
|
33 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"decoder_start_token_id": 0,
|
5 |
+
"eos_token_id": 2,
|
6 |
+
"pad_token_id": 0,
|
7 |
+
"transformers_version": "4.38.2"
|
8 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4e16bb65606db1c6360d1407bf4f9d05ebfc1f8e963f1e94ba2af39276066c75
|
3 |
+
size 891635790
|
special_tokens_map.json
ADDED
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<FCTX>",
|
4 |
+
"</FCTX>",
|
5 |
+
"<ECTX>",
|
6 |
+
"</ECTX>",
|
7 |
+
"<PRIVATE_FCTX>",
|
8 |
+
"</PRIVATE_FCTX>"
|
9 |
+
],
|
10 |
+
"bos_token": {
|
11 |
+
"content": "<s>",
|
12 |
+
"lstrip": false,
|
13 |
+
"normalized": true,
|
14 |
+
"rstrip": false,
|
15 |
+
"single_word": false
|
16 |
+
},
|
17 |
+
"cls_token": {
|
18 |
+
"content": "<s>",
|
19 |
+
"lstrip": false,
|
20 |
+
"normalized": true,
|
21 |
+
"rstrip": false,
|
22 |
+
"single_word": false
|
23 |
+
},
|
24 |
+
"eos_token": {
|
25 |
+
"content": "</s>",
|
26 |
+
"lstrip": false,
|
27 |
+
"normalized": true,
|
28 |
+
"rstrip": false,
|
29 |
+
"single_word": false
|
30 |
+
},
|
31 |
+
"mask_token": {
|
32 |
+
"content": "<mask>",
|
33 |
+
"lstrip": true,
|
34 |
+
"normalized": true,
|
35 |
+
"rstrip": false,
|
36 |
+
"single_word": false
|
37 |
+
},
|
38 |
+
"pad_token": {
|
39 |
+
"content": "<pad>",
|
40 |
+
"lstrip": false,
|
41 |
+
"normalized": true,
|
42 |
+
"rstrip": false,
|
43 |
+
"single_word": false
|
44 |
+
},
|
45 |
+
"sep_token": {
|
46 |
+
"content": "</s>",
|
47 |
+
"lstrip": false,
|
48 |
+
"normalized": true,
|
49 |
+
"rstrip": false,
|
50 |
+
"single_word": false
|
51 |
+
},
|
52 |
+
"unk_token": {
|
53 |
+
"content": "<unk>",
|
54 |
+
"lstrip": false,
|
55 |
+
"normalized": true,
|
56 |
+
"rstrip": false,
|
57 |
+
"single_word": false
|
58 |
+
}
|
59 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"0": {
|
5 |
+
"content": "<pad>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
},
|
12 |
+
"1": {
|
13 |
+
"content": "<s>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": true,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": true
|
19 |
+
},
|
20 |
+
"2": {
|
21 |
+
"content": "</s>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": true,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false,
|
26 |
+
"special": true
|
27 |
+
},
|
28 |
+
"3": {
|
29 |
+
"content": "<unk>",
|
30 |
+
"lstrip": false,
|
31 |
+
"normalized": true,
|
32 |
+
"rstrip": false,
|
33 |
+
"single_word": false,
|
34 |
+
"special": true
|
35 |
+
},
|
36 |
+
"4": {
|
37 |
+
"content": "<mask>",
|
38 |
+
"lstrip": true,
|
39 |
+
"normalized": true,
|
40 |
+
"rstrip": false,
|
41 |
+
"single_word": false,
|
42 |
+
"special": true
|
43 |
+
},
|
44 |
+
"32100": {
|
45 |
+
"content": "<FCTX>",
|
46 |
+
"lstrip": false,
|
47 |
+
"normalized": false,
|
48 |
+
"rstrip": false,
|
49 |
+
"single_word": false,
|
50 |
+
"special": true
|
51 |
+
},
|
52 |
+
"32101": {
|
53 |
+
"content": "</FCTX>",
|
54 |
+
"lstrip": false,
|
55 |
+
"normalized": false,
|
56 |
+
"rstrip": false,
|
57 |
+
"single_word": false,
|
58 |
+
"special": true
|
59 |
+
},
|
60 |
+
"32102": {
|
61 |
+
"content": "<ECTX>",
|
62 |
+
"lstrip": false,
|
63 |
+
"normalized": false,
|
64 |
+
"rstrip": false,
|
65 |
+
"single_word": false,
|
66 |
+
"special": true
|
67 |
+
},
|
68 |
+
"32103": {
|
69 |
+
"content": "</ECTX>",
|
70 |
+
"lstrip": false,
|
71 |
+
"normalized": false,
|
72 |
+
"rstrip": false,
|
73 |
+
"single_word": false,
|
74 |
+
"special": true
|
75 |
+
},
|
76 |
+
"32104": {
|
77 |
+
"content": "<PRIVATE_FCTX>",
|
78 |
+
"lstrip": false,
|
79 |
+
"normalized": false,
|
80 |
+
"rstrip": false,
|
81 |
+
"single_word": false,
|
82 |
+
"special": true
|
83 |
+
},
|
84 |
+
"32105": {
|
85 |
+
"content": "</PRIVATE_FCTX>",
|
86 |
+
"lstrip": false,
|
87 |
+
"normalized": false,
|
88 |
+
"rstrip": false,
|
89 |
+
"single_word": false,
|
90 |
+
"special": true
|
91 |
+
}
|
92 |
+
},
|
93 |
+
"additional_special_tokens": [
|
94 |
+
"<FCTX>",
|
95 |
+
"</FCTX>",
|
96 |
+
"<ECTX>",
|
97 |
+
"</ECTX>",
|
98 |
+
"<PRIVATE_FCTX>",
|
99 |
+
"</PRIVATE_FCTX>"
|
100 |
+
],
|
101 |
+
"bos_token": "<s>",
|
102 |
+
"clean_up_tokenization_spaces": true,
|
103 |
+
"cls_token": "<s>",
|
104 |
+
"eos_token": "</s>",
|
105 |
+
"errors": "replace",
|
106 |
+
"mask_token": "<mask>",
|
107 |
+
"model_max_length": 512,
|
108 |
+
"pad_token": "<pad>",
|
109 |
+
"sep_token": "</s>",
|
110 |
+
"tokenizer_class": "RobertaTokenizer",
|
111 |
+
"trim_offsets": true,
|
112 |
+
"unk_token": "<unk>"
|
113 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1603 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.0747156143188477,
|
3 |
+
"best_model_checkpoint": "/root/finetuning_executions/finetuning_01_codet5p_src_fm_fc_dctx/checkpoint-17548",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 87740,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.02,
|
13 |
+
"grad_norm": 1.5826328992843628,
|
14 |
+
"learning_rate": 2.4750000000000002e-05,
|
15 |
+
"loss": 1.404,
|
16 |
+
"step": 400
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.05,
|
20 |
+
"grad_norm": 1.5471534729003906,
|
21 |
+
"learning_rate": 4.975e-05,
|
22 |
+
"loss": 1.1707,
|
23 |
+
"step": 800
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.07,
|
27 |
+
"grad_norm": 1.150067687034607,
|
28 |
+
"learning_rate": 4.977225672877847e-05,
|
29 |
+
"loss": 1.1319,
|
30 |
+
"step": 1200
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.09,
|
34 |
+
"grad_norm": 1.197001338005066,
|
35 |
+
"learning_rate": 4.9542213020473895e-05,
|
36 |
+
"loss": 1.1168,
|
37 |
+
"step": 1600
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.11,
|
41 |
+
"grad_norm": 1.0678050518035889,
|
42 |
+
"learning_rate": 4.931216931216932e-05,
|
43 |
+
"loss": 1.098,
|
44 |
+
"step": 2000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.14,
|
48 |
+
"grad_norm": 1.196992039680481,
|
49 |
+
"learning_rate": 4.9082125603864734e-05,
|
50 |
+
"loss": 1.0811,
|
51 |
+
"step": 2400
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.16,
|
55 |
+
"grad_norm": 1.0403153896331787,
|
56 |
+
"learning_rate": 4.885208189556016e-05,
|
57 |
+
"loss": 1.0642,
|
58 |
+
"step": 2800
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.18,
|
62 |
+
"grad_norm": 1.290574073791504,
|
63 |
+
"learning_rate": 4.862203818725558e-05,
|
64 |
+
"loss": 1.0573,
|
65 |
+
"step": 3200
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.21,
|
69 |
+
"grad_norm": 1.1538596153259277,
|
70 |
+
"learning_rate": 4.8391994478951e-05,
|
71 |
+
"loss": 1.0422,
|
72 |
+
"step": 3600
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.23,
|
76 |
+
"grad_norm": 1.1622496843338013,
|
77 |
+
"learning_rate": 4.8161950770646426e-05,
|
78 |
+
"loss": 1.0302,
|
79 |
+
"step": 4000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.25,
|
83 |
+
"grad_norm": 1.2310576438903809,
|
84 |
+
"learning_rate": 4.793190706234185e-05,
|
85 |
+
"loss": 1.0327,
|
86 |
+
"step": 4400
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.27,
|
90 |
+
"grad_norm": 1.7041934728622437,
|
91 |
+
"learning_rate": 4.770186335403727e-05,
|
92 |
+
"loss": 1.016,
|
93 |
+
"step": 4800
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.3,
|
97 |
+
"grad_norm": 1.2586933374404907,
|
98 |
+
"learning_rate": 4.747181964573269e-05,
|
99 |
+
"loss": 1.022,
|
100 |
+
"step": 5200
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.32,
|
104 |
+
"grad_norm": 0.9497622847557068,
|
105 |
+
"learning_rate": 4.724177593742811e-05,
|
106 |
+
"loss": 1.0008,
|
107 |
+
"step": 5600
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.34,
|
111 |
+
"grad_norm": 1.1111699342727661,
|
112 |
+
"learning_rate": 4.7011732229123534e-05,
|
113 |
+
"loss": 0.9878,
|
114 |
+
"step": 6000
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.36,
|
118 |
+
"grad_norm": 1.177351951599121,
|
119 |
+
"learning_rate": 4.678168852081896e-05,
|
120 |
+
"loss": 0.9804,
|
121 |
+
"step": 6400
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.39,
|
125 |
+
"grad_norm": 1.1454102993011475,
|
126 |
+
"learning_rate": 4.655164481251438e-05,
|
127 |
+
"loss": 0.9739,
|
128 |
+
"step": 6800
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.41,
|
132 |
+
"grad_norm": 1.0769121646881104,
|
133 |
+
"learning_rate": 4.63216011042098e-05,
|
134 |
+
"loss": 0.9873,
|
135 |
+
"step": 7200
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.43,
|
139 |
+
"grad_norm": 1.2958855628967285,
|
140 |
+
"learning_rate": 4.609155739590522e-05,
|
141 |
+
"loss": 0.9615,
|
142 |
+
"step": 7600
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.46,
|
146 |
+
"grad_norm": 1.1914528608322144,
|
147 |
+
"learning_rate": 4.586151368760065e-05,
|
148 |
+
"loss": 0.9637,
|
149 |
+
"step": 8000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.48,
|
153 |
+
"grad_norm": 1.036251425743103,
|
154 |
+
"learning_rate": 4.563146997929607e-05,
|
155 |
+
"loss": 0.9651,
|
156 |
+
"step": 8400
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.5,
|
160 |
+
"grad_norm": 1.2706526517868042,
|
161 |
+
"learning_rate": 4.5401426270991495e-05,
|
162 |
+
"loss": 0.9538,
|
163 |
+
"step": 8800
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.52,
|
167 |
+
"grad_norm": 1.1763725280761719,
|
168 |
+
"learning_rate": 4.517138256268692e-05,
|
169 |
+
"loss": 0.9481,
|
170 |
+
"step": 9200
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.55,
|
174 |
+
"grad_norm": 1.0269743204116821,
|
175 |
+
"learning_rate": 4.4941338854382334e-05,
|
176 |
+
"loss": 0.9406,
|
177 |
+
"step": 9600
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.57,
|
181 |
+
"grad_norm": 1.1849923133850098,
|
182 |
+
"learning_rate": 4.471129514607776e-05,
|
183 |
+
"loss": 0.9325,
|
184 |
+
"step": 10000
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.59,
|
188 |
+
"grad_norm": 1.289588451385498,
|
189 |
+
"learning_rate": 4.448125143777318e-05,
|
190 |
+
"loss": 0.9198,
|
191 |
+
"step": 10400
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.62,
|
195 |
+
"grad_norm": 1.0682235956192017,
|
196 |
+
"learning_rate": 4.42512077294686e-05,
|
197 |
+
"loss": 0.938,
|
198 |
+
"step": 10800
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.64,
|
202 |
+
"grad_norm": 1.0467722415924072,
|
203 |
+
"learning_rate": 4.4021164021164026e-05,
|
204 |
+
"loss": 0.9191,
|
205 |
+
"step": 11200
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.66,
|
209 |
+
"grad_norm": 1.0143468379974365,
|
210 |
+
"learning_rate": 4.379112031285945e-05,
|
211 |
+
"loss": 0.9193,
|
212 |
+
"step": 11600
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.68,
|
216 |
+
"grad_norm": 1.0179190635681152,
|
217 |
+
"learning_rate": 4.3561076604554865e-05,
|
218 |
+
"loss": 0.9109,
|
219 |
+
"step": 12000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.71,
|
223 |
+
"grad_norm": 1.0898582935333252,
|
224 |
+
"learning_rate": 4.333103289625029e-05,
|
225 |
+
"loss": 0.9173,
|
226 |
+
"step": 12400
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.73,
|
230 |
+
"grad_norm": 1.0596779584884644,
|
231 |
+
"learning_rate": 4.310098918794571e-05,
|
232 |
+
"loss": 0.9236,
|
233 |
+
"step": 12800
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.75,
|
237 |
+
"grad_norm": 1.1058505773544312,
|
238 |
+
"learning_rate": 4.2870945479641134e-05,
|
239 |
+
"loss": 0.9006,
|
240 |
+
"step": 13200
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.78,
|
244 |
+
"grad_norm": 1.2219908237457275,
|
245 |
+
"learning_rate": 4.264090177133656e-05,
|
246 |
+
"loss": 0.9,
|
247 |
+
"step": 13600
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.8,
|
251 |
+
"grad_norm": 0.9692860245704651,
|
252 |
+
"learning_rate": 4.241143317230274e-05,
|
253 |
+
"loss": 0.9014,
|
254 |
+
"step": 14000
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.82,
|
258 |
+
"grad_norm": 1.1939465999603271,
|
259 |
+
"learning_rate": 4.218138946399816e-05,
|
260 |
+
"loss": 0.8905,
|
261 |
+
"step": 14400
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.84,
|
265 |
+
"grad_norm": 1.2843443155288696,
|
266 |
+
"learning_rate": 4.1951345755693586e-05,
|
267 |
+
"loss": 0.8988,
|
268 |
+
"step": 14800
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.87,
|
272 |
+
"grad_norm": 1.074813723564148,
|
273 |
+
"learning_rate": 4.1721302047389e-05,
|
274 |
+
"loss": 0.8979,
|
275 |
+
"step": 15200
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.89,
|
279 |
+
"grad_norm": 0.9322890043258667,
|
280 |
+
"learning_rate": 4.1491258339084425e-05,
|
281 |
+
"loss": 0.8782,
|
282 |
+
"step": 15600
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.91,
|
286 |
+
"grad_norm": 1.005053162574768,
|
287 |
+
"learning_rate": 4.126121463077985e-05,
|
288 |
+
"loss": 0.8735,
|
289 |
+
"step": 16000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.93,
|
293 |
+
"grad_norm": 1.0346530675888062,
|
294 |
+
"learning_rate": 4.103117092247527e-05,
|
295 |
+
"loss": 0.884,
|
296 |
+
"step": 16400
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.96,
|
300 |
+
"grad_norm": 1.113724946975708,
|
301 |
+
"learning_rate": 4.0801127214170694e-05,
|
302 |
+
"loss": 0.8765,
|
303 |
+
"step": 16800
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.98,
|
307 |
+
"grad_norm": 0.9634864926338196,
|
308 |
+
"learning_rate": 4.0571083505866117e-05,
|
309 |
+
"loss": 0.8734,
|
310 |
+
"step": 17200
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 1.0,
|
314 |
+
"eval_loss": 1.0747156143188477,
|
315 |
+
"eval_runtime": 239.6824,
|
316 |
+
"eval_samples_per_second": 251.479,
|
317 |
+
"eval_steps_per_second": 3.93,
|
318 |
+
"step": 17548
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 1.0,
|
322 |
+
"grad_norm": 0.9827917218208313,
|
323 |
+
"learning_rate": 4.034103979756153e-05,
|
324 |
+
"loss": 0.8706,
|
325 |
+
"step": 17600
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 1.03,
|
329 |
+
"grad_norm": 1.1454176902770996,
|
330 |
+
"learning_rate": 4.0110996089256956e-05,
|
331 |
+
"loss": 0.825,
|
332 |
+
"step": 18000
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.05,
|
336 |
+
"grad_norm": 0.9637428522109985,
|
337 |
+
"learning_rate": 3.988152749022314e-05,
|
338 |
+
"loss": 0.829,
|
339 |
+
"step": 18400
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.07,
|
343 |
+
"grad_norm": 1.0976300239562988,
|
344 |
+
"learning_rate": 3.965148378191857e-05,
|
345 |
+
"loss": 0.8276,
|
346 |
+
"step": 18800
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.09,
|
350 |
+
"grad_norm": 1.104611873626709,
|
351 |
+
"learning_rate": 3.9422015182884744e-05,
|
352 |
+
"loss": 0.8206,
|
353 |
+
"step": 19200
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.12,
|
357 |
+
"grad_norm": 1.373989224433899,
|
358 |
+
"learning_rate": 3.9193121693121694e-05,
|
359 |
+
"loss": 0.8225,
|
360 |
+
"step": 19600
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.14,
|
364 |
+
"grad_norm": 1.0706731081008911,
|
365 |
+
"learning_rate": 3.896307798481712e-05,
|
366 |
+
"loss": 0.8084,
|
367 |
+
"step": 20000
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 1.16,
|
371 |
+
"grad_norm": 1.1869192123413086,
|
372 |
+
"learning_rate": 3.873303427651253e-05,
|
373 |
+
"loss": 0.8222,
|
374 |
+
"step": 20400
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.19,
|
378 |
+
"grad_norm": 1.2390427589416504,
|
379 |
+
"learning_rate": 3.850299056820796e-05,
|
380 |
+
"loss": 0.8231,
|
381 |
+
"step": 20800
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 1.21,
|
385 |
+
"grad_norm": 1.0254757404327393,
|
386 |
+
"learning_rate": 3.8272946859903386e-05,
|
387 |
+
"loss": 0.8039,
|
388 |
+
"step": 21200
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 1.23,
|
392 |
+
"grad_norm": 1.2318751811981201,
|
393 |
+
"learning_rate": 3.804290315159881e-05,
|
394 |
+
"loss": 0.8122,
|
395 |
+
"step": 21600
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 1.25,
|
399 |
+
"grad_norm": 1.0611391067504883,
|
400 |
+
"learning_rate": 3.781285944329423e-05,
|
401 |
+
"loss": 0.8054,
|
402 |
+
"step": 22000
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 1.28,
|
406 |
+
"grad_norm": 1.1104662418365479,
|
407 |
+
"learning_rate": 3.7582815734989655e-05,
|
408 |
+
"loss": 0.8065,
|
409 |
+
"step": 22400
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 1.3,
|
413 |
+
"grad_norm": 0.9983170032501221,
|
414 |
+
"learning_rate": 3.735277202668507e-05,
|
415 |
+
"loss": 0.8058,
|
416 |
+
"step": 22800
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 1.32,
|
420 |
+
"grad_norm": 1.5960198640823364,
|
421 |
+
"learning_rate": 3.7122728318380494e-05,
|
422 |
+
"loss": 0.8135,
|
423 |
+
"step": 23200
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 1.34,
|
427 |
+
"grad_norm": 1.1167716979980469,
|
428 |
+
"learning_rate": 3.689268461007592e-05,
|
429 |
+
"loss": 0.7942,
|
430 |
+
"step": 23600
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 1.37,
|
434 |
+
"grad_norm": 1.1541776657104492,
|
435 |
+
"learning_rate": 3.666264090177134e-05,
|
436 |
+
"loss": 0.7997,
|
437 |
+
"step": 24000
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 1.39,
|
441 |
+
"grad_norm": 1.0213623046875,
|
442 |
+
"learning_rate": 3.643259719346676e-05,
|
443 |
+
"loss": 0.8009,
|
444 |
+
"step": 24400
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 1.41,
|
448 |
+
"grad_norm": 1.2691611051559448,
|
449 |
+
"learning_rate": 3.6202553485162186e-05,
|
450 |
+
"loss": 0.801,
|
451 |
+
"step": 24800
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 1.44,
|
455 |
+
"grad_norm": 1.0916258096694946,
|
456 |
+
"learning_rate": 3.59725097768576e-05,
|
457 |
+
"loss": 0.7868,
|
458 |
+
"step": 25200
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.46,
|
462 |
+
"grad_norm": 1.2260217666625977,
|
463 |
+
"learning_rate": 3.5742466068553025e-05,
|
464 |
+
"loss": 0.7905,
|
465 |
+
"step": 25600
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 1.48,
|
469 |
+
"grad_norm": 1.0053335428237915,
|
470 |
+
"learning_rate": 3.5513572578789974e-05,
|
471 |
+
"loss": 0.7976,
|
472 |
+
"step": 26000
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 1.5,
|
476 |
+
"grad_norm": 1.2022624015808105,
|
477 |
+
"learning_rate": 3.528352887048539e-05,
|
478 |
+
"loss": 0.7821,
|
479 |
+
"step": 26400
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.53,
|
483 |
+
"grad_norm": 1.7201203107833862,
|
484 |
+
"learning_rate": 3.5053485162180814e-05,
|
485 |
+
"loss": 0.7812,
|
486 |
+
"step": 26800
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 1.55,
|
490 |
+
"grad_norm": 1.288846731185913,
|
491 |
+
"learning_rate": 3.482344145387624e-05,
|
492 |
+
"loss": 0.7773,
|
493 |
+
"step": 27200
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 1.57,
|
497 |
+
"grad_norm": 0.9436947107315063,
|
498 |
+
"learning_rate": 3.459339774557166e-05,
|
499 |
+
"loss": 0.7816,
|
500 |
+
"step": 27600
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 1.6,
|
504 |
+
"grad_norm": 1.2146368026733398,
|
505 |
+
"learning_rate": 3.436335403726708e-05,
|
506 |
+
"loss": 0.7879,
|
507 |
+
"step": 28000
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 1.62,
|
511 |
+
"grad_norm": 1.2685799598693848,
|
512 |
+
"learning_rate": 3.4133310328962506e-05,
|
513 |
+
"loss": 0.7851,
|
514 |
+
"step": 28400
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 1.64,
|
518 |
+
"grad_norm": 1.0855278968811035,
|
519 |
+
"learning_rate": 3.390326662065793e-05,
|
520 |
+
"loss": 0.7878,
|
521 |
+
"step": 28800
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 1.66,
|
525 |
+
"grad_norm": 1.039978265762329,
|
526 |
+
"learning_rate": 3.367379802162411e-05,
|
527 |
+
"loss": 0.7842,
|
528 |
+
"step": 29200
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 1.69,
|
532 |
+
"grad_norm": 1.144707202911377,
|
533 |
+
"learning_rate": 3.344375431331953e-05,
|
534 |
+
"loss": 0.7835,
|
535 |
+
"step": 29600
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 1.71,
|
539 |
+
"grad_norm": 1.112640380859375,
|
540 |
+
"learning_rate": 3.321371060501495e-05,
|
541 |
+
"loss": 0.7787,
|
542 |
+
"step": 30000
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 1.73,
|
546 |
+
"grad_norm": 0.9749494194984436,
|
547 |
+
"learning_rate": 3.2983666896710374e-05,
|
548 |
+
"loss": 0.7665,
|
549 |
+
"step": 30400
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 1.76,
|
553 |
+
"grad_norm": 1.2835110425949097,
|
554 |
+
"learning_rate": 3.2753623188405796e-05,
|
555 |
+
"loss": 0.7778,
|
556 |
+
"step": 30800
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 1.78,
|
560 |
+
"grad_norm": 1.1826539039611816,
|
561 |
+
"learning_rate": 3.252357948010122e-05,
|
562 |
+
"loss": 0.7702,
|
563 |
+
"step": 31200
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 1.8,
|
567 |
+
"grad_norm": 1.0026845932006836,
|
568 |
+
"learning_rate": 3.229353577179664e-05,
|
569 |
+
"loss": 0.7807,
|
570 |
+
"step": 31600
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 1.82,
|
574 |
+
"grad_norm": 1.314173698425293,
|
575 |
+
"learning_rate": 3.2063492063492065e-05,
|
576 |
+
"loss": 0.7693,
|
577 |
+
"step": 32000
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 1.85,
|
581 |
+
"grad_norm": 1.2866084575653076,
|
582 |
+
"learning_rate": 3.183344835518749e-05,
|
583 |
+
"loss": 0.7672,
|
584 |
+
"step": 32400
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 1.87,
|
588 |
+
"grad_norm": 1.2260661125183105,
|
589 |
+
"learning_rate": 3.160340464688291e-05,
|
590 |
+
"loss": 0.7679,
|
591 |
+
"step": 32800
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 1.89,
|
595 |
+
"grad_norm": 1.05460524559021,
|
596 |
+
"learning_rate": 3.1373360938578334e-05,
|
597 |
+
"loss": 0.7678,
|
598 |
+
"step": 33200
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 1.91,
|
602 |
+
"grad_norm": 1.020358920097351,
|
603 |
+
"learning_rate": 3.114331723027376e-05,
|
604 |
+
"loss": 0.7543,
|
605 |
+
"step": 33600
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 1.94,
|
609 |
+
"grad_norm": 1.2354198694229126,
|
610 |
+
"learning_rate": 3.091327352196918e-05,
|
611 |
+
"loss": 0.7627,
|
612 |
+
"step": 34000
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 1.96,
|
616 |
+
"grad_norm": 1.1465234756469727,
|
617 |
+
"learning_rate": 3.0683229813664596e-05,
|
618 |
+
"loss": 0.7575,
|
619 |
+
"step": 34400
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 1.98,
|
623 |
+
"grad_norm": 1.2132583856582642,
|
624 |
+
"learning_rate": 3.045318610536002e-05,
|
625 |
+
"loss": 0.7667,
|
626 |
+
"step": 34800
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 2.0,
|
630 |
+
"eval_loss": 1.0832360982894897,
|
631 |
+
"eval_runtime": 239.646,
|
632 |
+
"eval_samples_per_second": 251.517,
|
633 |
+
"eval_steps_per_second": 3.931,
|
634 |
+
"step": 35096
|
635 |
+
},
|
636 |
+
{
|
637 |
+
"epoch": 2.01,
|
638 |
+
"grad_norm": 1.150380253791809,
|
639 |
+
"learning_rate": 3.0223142397055442e-05,
|
640 |
+
"loss": 0.7454,
|
641 |
+
"step": 35200
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 2.03,
|
645 |
+
"grad_norm": 1.1517704725265503,
|
646 |
+
"learning_rate": 2.9993098688750865e-05,
|
647 |
+
"loss": 0.7151,
|
648 |
+
"step": 35600
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 2.05,
|
652 |
+
"grad_norm": 1.2216242551803589,
|
653 |
+
"learning_rate": 2.9763054980446285e-05,
|
654 |
+
"loss": 0.7085,
|
655 |
+
"step": 36000
|
656 |
+
},
|
657 |
+
{
|
658 |
+
"epoch": 2.07,
|
659 |
+
"grad_norm": 1.1124303340911865,
|
660 |
+
"learning_rate": 2.9533011272141708e-05,
|
661 |
+
"loss": 0.7112,
|
662 |
+
"step": 36400
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 2.1,
|
666 |
+
"grad_norm": 1.2349203824996948,
|
667 |
+
"learning_rate": 2.930296756383713e-05,
|
668 |
+
"loss": 0.7028,
|
669 |
+
"step": 36800
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 2.12,
|
673 |
+
"grad_norm": 1.1383484601974487,
|
674 |
+
"learning_rate": 2.9073498964803314e-05,
|
675 |
+
"loss": 0.7043,
|
676 |
+
"step": 37200
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 2.14,
|
680 |
+
"grad_norm": 1.156599521636963,
|
681 |
+
"learning_rate": 2.8843455256498737e-05,
|
682 |
+
"loss": 0.7171,
|
683 |
+
"step": 37600
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 2.17,
|
687 |
+
"grad_norm": 1.2010080814361572,
|
688 |
+
"learning_rate": 2.8613411548194156e-05,
|
689 |
+
"loss": 0.7131,
|
690 |
+
"step": 38000
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 2.19,
|
694 |
+
"grad_norm": 1.0973454713821411,
|
695 |
+
"learning_rate": 2.838336783988958e-05,
|
696 |
+
"loss": 0.7107,
|
697 |
+
"step": 38400
|
698 |
+
},
|
699 |
+
{
|
700 |
+
"epoch": 2.21,
|
701 |
+
"grad_norm": 1.1451126337051392,
|
702 |
+
"learning_rate": 2.8153324131585002e-05,
|
703 |
+
"loss": 0.7144,
|
704 |
+
"step": 38800
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 2.23,
|
708 |
+
"grad_norm": 1.3358917236328125,
|
709 |
+
"learning_rate": 2.7923855532551185e-05,
|
710 |
+
"loss": 0.7095,
|
711 |
+
"step": 39200
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 2.26,
|
715 |
+
"grad_norm": 1.142595887184143,
|
716 |
+
"learning_rate": 2.7694386933517368e-05,
|
717 |
+
"loss": 0.7057,
|
718 |
+
"step": 39600
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 2.28,
|
722 |
+
"grad_norm": 1.0060242414474487,
|
723 |
+
"learning_rate": 2.746434322521279e-05,
|
724 |
+
"loss": 0.7015,
|
725 |
+
"step": 40000
|
726 |
+
},
|
727 |
+
{
|
728 |
+
"epoch": 2.3,
|
729 |
+
"grad_norm": 0.9941455721855164,
|
730 |
+
"learning_rate": 2.7234299516908214e-05,
|
731 |
+
"loss": 0.7132,
|
732 |
+
"step": 40400
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 2.33,
|
736 |
+
"grad_norm": 1.0412284135818481,
|
737 |
+
"learning_rate": 2.7004255808603634e-05,
|
738 |
+
"loss": 0.705,
|
739 |
+
"step": 40800
|
740 |
+
},
|
741 |
+
{
|
742 |
+
"epoch": 2.35,
|
743 |
+
"grad_norm": 1.099241852760315,
|
744 |
+
"learning_rate": 2.6774212100299057e-05,
|
745 |
+
"loss": 0.7084,
|
746 |
+
"step": 41200
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 2.37,
|
750 |
+
"grad_norm": 1.1212828159332275,
|
751 |
+
"learning_rate": 2.654474350126524e-05,
|
752 |
+
"loss": 0.6978,
|
753 |
+
"step": 41600
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 2.39,
|
757 |
+
"grad_norm": 1.0526177883148193,
|
758 |
+
"learning_rate": 2.6315274902231422e-05,
|
759 |
+
"loss": 0.7004,
|
760 |
+
"step": 42000
|
761 |
+
},
|
762 |
+
{
|
763 |
+
"epoch": 2.42,
|
764 |
+
"grad_norm": 1.1355630159378052,
|
765 |
+
"learning_rate": 2.6085231193926845e-05,
|
766 |
+
"loss": 0.7066,
|
767 |
+
"step": 42400
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 2.44,
|
771 |
+
"grad_norm": 1.1076544523239136,
|
772 |
+
"learning_rate": 2.5855187485622268e-05,
|
773 |
+
"loss": 0.7059,
|
774 |
+
"step": 42800
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 2.46,
|
778 |
+
"grad_norm": 0.9984206557273865,
|
779 |
+
"learning_rate": 2.562514377731769e-05,
|
780 |
+
"loss": 0.6959,
|
781 |
+
"step": 43200
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 2.48,
|
785 |
+
"grad_norm": 1.267728567123413,
|
786 |
+
"learning_rate": 2.5395675178283874e-05,
|
787 |
+
"loss": 0.7109,
|
788 |
+
"step": 43600
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 2.51,
|
792 |
+
"grad_norm": 1.010473370552063,
|
793 |
+
"learning_rate": 2.5165631469979294e-05,
|
794 |
+
"loss": 0.707,
|
795 |
+
"step": 44000
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 2.53,
|
799 |
+
"grad_norm": 1.3132517337799072,
|
800 |
+
"learning_rate": 2.493616287094548e-05,
|
801 |
+
"loss": 0.7049,
|
802 |
+
"step": 44400
|
803 |
+
},
|
804 |
+
{
|
805 |
+
"epoch": 2.55,
|
806 |
+
"grad_norm": 1.0806896686553955,
|
807 |
+
"learning_rate": 2.4706119162640903e-05,
|
808 |
+
"loss": 0.7032,
|
809 |
+
"step": 44800
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 2.58,
|
813 |
+
"grad_norm": 1.0918660163879395,
|
814 |
+
"learning_rate": 2.4476075454336326e-05,
|
815 |
+
"loss": 0.6876,
|
816 |
+
"step": 45200
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 2.6,
|
820 |
+
"grad_norm": 1.172402024269104,
|
821 |
+
"learning_rate": 2.424603174603175e-05,
|
822 |
+
"loss": 0.697,
|
823 |
+
"step": 45600
|
824 |
+
},
|
825 |
+
{
|
826 |
+
"epoch": 2.62,
|
827 |
+
"grad_norm": 1.3677465915679932,
|
828 |
+
"learning_rate": 2.4015988037727168e-05,
|
829 |
+
"loss": 0.694,
|
830 |
+
"step": 46000
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 2.64,
|
834 |
+
"grad_norm": 0.9764865636825562,
|
835 |
+
"learning_rate": 2.378594432942259e-05,
|
836 |
+
"loss": 0.692,
|
837 |
+
"step": 46400
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 2.67,
|
841 |
+
"grad_norm": 1.168467402458191,
|
842 |
+
"learning_rate": 2.3555900621118014e-05,
|
843 |
+
"loss": 0.6907,
|
844 |
+
"step": 46800
|
845 |
+
},
|
846 |
+
{
|
847 |
+
"epoch": 2.69,
|
848 |
+
"grad_norm": 1.0712902545928955,
|
849 |
+
"learning_rate": 2.3325856912813434e-05,
|
850 |
+
"loss": 0.6969,
|
851 |
+
"step": 47200
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 2.71,
|
855 |
+
"grad_norm": 1.2307454347610474,
|
856 |
+
"learning_rate": 2.3095813204508857e-05,
|
857 |
+
"loss": 0.6912,
|
858 |
+
"step": 47600
|
859 |
+
},
|
860 |
+
{
|
861 |
+
"epoch": 2.74,
|
862 |
+
"grad_norm": 1.0218665599822998,
|
863 |
+
"learning_rate": 2.286576949620428e-05,
|
864 |
+
"loss": 0.6977,
|
865 |
+
"step": 48000
|
866 |
+
},
|
867 |
+
{
|
868 |
+
"epoch": 2.76,
|
869 |
+
"grad_norm": 1.2914952039718628,
|
870 |
+
"learning_rate": 2.2635725787899703e-05,
|
871 |
+
"loss": 0.6902,
|
872 |
+
"step": 48400
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 2.78,
|
876 |
+
"grad_norm": 1.1566945314407349,
|
877 |
+
"learning_rate": 2.2405682079595126e-05,
|
878 |
+
"loss": 0.6812,
|
879 |
+
"step": 48800
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 2.8,
|
883 |
+
"grad_norm": 1.0954487323760986,
|
884 |
+
"learning_rate": 2.2175638371290545e-05,
|
885 |
+
"loss": 0.6884,
|
886 |
+
"step": 49200
|
887 |
+
},
|
888 |
+
{
|
889 |
+
"epoch": 2.83,
|
890 |
+
"grad_norm": 1.2148932218551636,
|
891 |
+
"learning_rate": 2.1945594662985968e-05,
|
892 |
+
"loss": 0.6815,
|
893 |
+
"step": 49600
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 2.85,
|
897 |
+
"grad_norm": 1.3392747640609741,
|
898 |
+
"learning_rate": 2.171555095468139e-05,
|
899 |
+
"loss": 0.68,
|
900 |
+
"step": 50000
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 2.87,
|
904 |
+
"grad_norm": 1.1132878065109253,
|
905 |
+
"learning_rate": 2.1485507246376814e-05,
|
906 |
+
"loss": 0.6985,
|
907 |
+
"step": 50400
|
908 |
+
},
|
909 |
+
{
|
910 |
+
"epoch": 2.89,
|
911 |
+
"grad_norm": 1.1941944360733032,
|
912 |
+
"learning_rate": 2.1255463538072234e-05,
|
913 |
+
"loss": 0.6888,
|
914 |
+
"step": 50800
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 2.92,
|
918 |
+
"grad_norm": 1.3875670433044434,
|
919 |
+
"learning_rate": 2.1025994939038417e-05,
|
920 |
+
"loss": 0.689,
|
921 |
+
"step": 51200
|
922 |
+
},
|
923 |
+
{
|
924 |
+
"epoch": 2.94,
|
925 |
+
"grad_norm": 1.1829801797866821,
|
926 |
+
"learning_rate": 2.079595123073384e-05,
|
927 |
+
"loss": 0.6913,
|
928 |
+
"step": 51600
|
929 |
+
},
|
930 |
+
{
|
931 |
+
"epoch": 2.96,
|
932 |
+
"grad_norm": 1.1661145687103271,
|
933 |
+
"learning_rate": 2.0565907522429263e-05,
|
934 |
+
"loss": 0.6889,
|
935 |
+
"step": 52000
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 2.99,
|
939 |
+
"grad_norm": 1.4086240530014038,
|
940 |
+
"learning_rate": 2.0335863814124682e-05,
|
941 |
+
"loss": 0.6835,
|
942 |
+
"step": 52400
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 3.0,
|
946 |
+
"eval_loss": 1.08446204662323,
|
947 |
+
"eval_runtime": 239.6214,
|
948 |
+
"eval_samples_per_second": 251.543,
|
949 |
+
"eval_steps_per_second": 3.931,
|
950 |
+
"step": 52644
|
951 |
+
},
|
952 |
+
{
|
953 |
+
"epoch": 3.01,
|
954 |
+
"grad_norm": 1.4694873094558716,
|
955 |
+
"learning_rate": 2.0105820105820105e-05,
|
956 |
+
"loss": 0.6682,
|
957 |
+
"step": 52800
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 3.03,
|
961 |
+
"grad_norm": 1.2044637203216553,
|
962 |
+
"learning_rate": 1.9875776397515528e-05,
|
963 |
+
"loss": 0.6507,
|
964 |
+
"step": 53200
|
965 |
+
},
|
966 |
+
{
|
967 |
+
"epoch": 3.05,
|
968 |
+
"grad_norm": 1.100796103477478,
|
969 |
+
"learning_rate": 1.964573268921095e-05,
|
970 |
+
"loss": 0.6401,
|
971 |
+
"step": 53600
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 3.08,
|
975 |
+
"grad_norm": 1.0444916486740112,
|
976 |
+
"learning_rate": 1.9415688980906374e-05,
|
977 |
+
"loss": 0.6449,
|
978 |
+
"step": 54000
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 3.1,
|
982 |
+
"grad_norm": 1.0386139154434204,
|
983 |
+
"learning_rate": 1.918679549114332e-05,
|
984 |
+
"loss": 0.6501,
|
985 |
+
"step": 54400
|
986 |
+
},
|
987 |
+
{
|
988 |
+
"epoch": 3.12,
|
989 |
+
"grad_norm": 1.1518653631210327,
|
990 |
+
"learning_rate": 1.895675178283874e-05,
|
991 |
+
"loss": 0.6461,
|
992 |
+
"step": 54800
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 3.15,
|
996 |
+
"grad_norm": 1.1994818449020386,
|
997 |
+
"learning_rate": 1.8726708074534163e-05,
|
998 |
+
"loss": 0.6556,
|
999 |
+
"step": 55200
|
1000 |
+
},
|
1001 |
+
{
|
1002 |
+
"epoch": 3.17,
|
1003 |
+
"grad_norm": 1.1442828178405762,
|
1004 |
+
"learning_rate": 1.8496664366229586e-05,
|
1005 |
+
"loss": 0.6361,
|
1006 |
+
"step": 55600
|
1007 |
+
},
|
1008 |
+
{
|
1009 |
+
"epoch": 3.19,
|
1010 |
+
"grad_norm": 1.0276410579681396,
|
1011 |
+
"learning_rate": 1.826662065792501e-05,
|
1012 |
+
"loss": 0.6378,
|
1013 |
+
"step": 56000
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 3.21,
|
1017 |
+
"grad_norm": 1.1680177450180054,
|
1018 |
+
"learning_rate": 1.8036576949620428e-05,
|
1019 |
+
"loss": 0.6439,
|
1020 |
+
"step": 56400
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 3.24,
|
1024 |
+
"grad_norm": 1.1519464254379272,
|
1025 |
+
"learning_rate": 1.780653324131585e-05,
|
1026 |
+
"loss": 0.6466,
|
1027 |
+
"step": 56800
|
1028 |
+
},
|
1029 |
+
{
|
1030 |
+
"epoch": 3.26,
|
1031 |
+
"grad_norm": 1.1728110313415527,
|
1032 |
+
"learning_rate": 1.7576489533011274e-05,
|
1033 |
+
"loss": 0.6419,
|
1034 |
+
"step": 57200
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 3.28,
|
1038 |
+
"grad_norm": 1.1479480266571045,
|
1039 |
+
"learning_rate": 1.7347020933977457e-05,
|
1040 |
+
"loss": 0.6398,
|
1041 |
+
"step": 57600
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 3.31,
|
1045 |
+
"grad_norm": 1.3310575485229492,
|
1046 |
+
"learning_rate": 1.7116977225672877e-05,
|
1047 |
+
"loss": 0.6407,
|
1048 |
+
"step": 58000
|
1049 |
+
},
|
1050 |
+
{
|
1051 |
+
"epoch": 3.33,
|
1052 |
+
"grad_norm": 1.3201924562454224,
|
1053 |
+
"learning_rate": 1.68869335173683e-05,
|
1054 |
+
"loss": 0.643,
|
1055 |
+
"step": 58400
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 3.35,
|
1059 |
+
"grad_norm": 1.1604301929473877,
|
1060 |
+
"learning_rate": 1.6656889809063723e-05,
|
1061 |
+
"loss": 0.638,
|
1062 |
+
"step": 58800
|
1063 |
+
},
|
1064 |
+
{
|
1065 |
+
"epoch": 3.37,
|
1066 |
+
"grad_norm": 1.0410959720611572,
|
1067 |
+
"learning_rate": 1.6426846100759146e-05,
|
1068 |
+
"loss": 0.6418,
|
1069 |
+
"step": 59200
|
1070 |
+
},
|
1071 |
+
{
|
1072 |
+
"epoch": 3.4,
|
1073 |
+
"grad_norm": 1.3607354164123535,
|
1074 |
+
"learning_rate": 1.6196802392454565e-05,
|
1075 |
+
"loss": 0.6486,
|
1076 |
+
"step": 59600
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 3.42,
|
1080 |
+
"grad_norm": 1.0975886583328247,
|
1081 |
+
"learning_rate": 1.596675868414999e-05,
|
1082 |
+
"loss": 0.6309,
|
1083 |
+
"step": 60000
|
1084 |
+
},
|
1085 |
+
{
|
1086 |
+
"epoch": 3.44,
|
1087 |
+
"grad_norm": 1.1079365015029907,
|
1088 |
+
"learning_rate": 1.573671497584541e-05,
|
1089 |
+
"loss": 0.6362,
|
1090 |
+
"step": 60400
|
1091 |
+
},
|
1092 |
+
{
|
1093 |
+
"epoch": 3.46,
|
1094 |
+
"grad_norm": 1.1985440254211426,
|
1095 |
+
"learning_rate": 1.5506671267540834e-05,
|
1096 |
+
"loss": 0.6402,
|
1097 |
+
"step": 60800
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 3.49,
|
1101 |
+
"grad_norm": 0.9779540300369263,
|
1102 |
+
"learning_rate": 1.5276627559236257e-05,
|
1103 |
+
"loss": 0.6441,
|
1104 |
+
"step": 61200
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 3.51,
|
1108 |
+
"grad_norm": 1.2140247821807861,
|
1109 |
+
"learning_rate": 1.5046583850931678e-05,
|
1110 |
+
"loss": 0.6393,
|
1111 |
+
"step": 61600
|
1112 |
+
},
|
1113 |
+
{
|
1114 |
+
"epoch": 3.53,
|
1115 |
+
"grad_norm": 1.3921316862106323,
|
1116 |
+
"learning_rate": 1.48165401426271e-05,
|
1117 |
+
"loss": 0.6407,
|
1118 |
+
"step": 62000
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 3.56,
|
1122 |
+
"grad_norm": 1.286074161529541,
|
1123 |
+
"learning_rate": 1.4587071543593284e-05,
|
1124 |
+
"loss": 0.6447,
|
1125 |
+
"step": 62400
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"epoch": 3.58,
|
1129 |
+
"grad_norm": 1.2328264713287354,
|
1130 |
+
"learning_rate": 1.4357027835288705e-05,
|
1131 |
+
"loss": 0.6387,
|
1132 |
+
"step": 62800
|
1133 |
+
},
|
1134 |
+
{
|
1135 |
+
"epoch": 3.6,
|
1136 |
+
"grad_norm": 1.1091105937957764,
|
1137 |
+
"learning_rate": 1.4126984126984127e-05,
|
1138 |
+
"loss": 0.6287,
|
1139 |
+
"step": 63200
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 3.62,
|
1143 |
+
"grad_norm": 1.3000015020370483,
|
1144 |
+
"learning_rate": 1.389694041867955e-05,
|
1145 |
+
"loss": 0.6421,
|
1146 |
+
"step": 63600
|
1147 |
+
},
|
1148 |
+
{
|
1149 |
+
"epoch": 3.65,
|
1150 |
+
"grad_norm": 0.9806800484657288,
|
1151 |
+
"learning_rate": 1.3666896710374971e-05,
|
1152 |
+
"loss": 0.6342,
|
1153 |
+
"step": 64000
|
1154 |
+
},
|
1155 |
+
{
|
1156 |
+
"epoch": 3.67,
|
1157 |
+
"grad_norm": 1.4063104391098022,
|
1158 |
+
"learning_rate": 1.3436853002070392e-05,
|
1159 |
+
"loss": 0.6323,
|
1160 |
+
"step": 64400
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 3.69,
|
1164 |
+
"grad_norm": 1.1137511730194092,
|
1165 |
+
"learning_rate": 1.3206809293765815e-05,
|
1166 |
+
"loss": 0.6316,
|
1167 |
+
"step": 64800
|
1168 |
+
},
|
1169 |
+
{
|
1170 |
+
"epoch": 3.72,
|
1171 |
+
"grad_norm": 1.307237148284912,
|
1172 |
+
"learning_rate": 1.2976765585461237e-05,
|
1173 |
+
"loss": 0.6209,
|
1174 |
+
"step": 65200
|
1175 |
+
},
|
1176 |
+
{
|
1177 |
+
"epoch": 3.74,
|
1178 |
+
"grad_norm": 1.2854503393173218,
|
1179 |
+
"learning_rate": 1.2746721877156661e-05,
|
1180 |
+
"loss": 0.635,
|
1181 |
+
"step": 65600
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 3.76,
|
1185 |
+
"grad_norm": 1.4092977046966553,
|
1186 |
+
"learning_rate": 1.2516678168852084e-05,
|
1187 |
+
"loss": 0.6337,
|
1188 |
+
"step": 66000
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 3.78,
|
1192 |
+
"grad_norm": 1.146622657775879,
|
1193 |
+
"learning_rate": 1.2286634460547504e-05,
|
1194 |
+
"loss": 0.621,
|
1195 |
+
"step": 66400
|
1196 |
+
},
|
1197 |
+
{
|
1198 |
+
"epoch": 3.81,
|
1199 |
+
"grad_norm": 1.4500105381011963,
|
1200 |
+
"learning_rate": 1.2056590752242927e-05,
|
1201 |
+
"loss": 0.6351,
|
1202 |
+
"step": 66800
|
1203 |
+
},
|
1204 |
+
{
|
1205 |
+
"epoch": 3.83,
|
1206 |
+
"grad_norm": 1.3374382257461548,
|
1207 |
+
"learning_rate": 1.182654704393835e-05,
|
1208 |
+
"loss": 0.6434,
|
1209 |
+
"step": 67200
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 3.85,
|
1213 |
+
"grad_norm": 1.172215223312378,
|
1214 |
+
"learning_rate": 1.1596503335633771e-05,
|
1215 |
+
"loss": 0.6299,
|
1216 |
+
"step": 67600
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 3.88,
|
1220 |
+
"grad_norm": 1.228962779045105,
|
1221 |
+
"learning_rate": 1.1366459627329192e-05,
|
1222 |
+
"loss": 0.6289,
|
1223 |
+
"step": 68000
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 3.9,
|
1227 |
+
"grad_norm": 1.1930328607559204,
|
1228 |
+
"learning_rate": 1.1136415919024615e-05,
|
1229 |
+
"loss": 0.6387,
|
1230 |
+
"step": 68400
|
1231 |
+
},
|
1232 |
+
{
|
1233 |
+
"epoch": 3.92,
|
1234 |
+
"grad_norm": 1.0659236907958984,
|
1235 |
+
"learning_rate": 1.0906947319990798e-05,
|
1236 |
+
"loss": 0.6323,
|
1237 |
+
"step": 68800
|
1238 |
+
},
|
1239 |
+
{
|
1240 |
+
"epoch": 3.94,
|
1241 |
+
"grad_norm": 1.2611548900604248,
|
1242 |
+
"learning_rate": 1.0677478720956983e-05,
|
1243 |
+
"loss": 0.6287,
|
1244 |
+
"step": 69200
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 3.97,
|
1248 |
+
"grad_norm": 1.186289668083191,
|
1249 |
+
"learning_rate": 1.0447435012652404e-05,
|
1250 |
+
"loss": 0.6232,
|
1251 |
+
"step": 69600
|
1252 |
+
},
|
1253 |
+
{
|
1254 |
+
"epoch": 3.99,
|
1255 |
+
"grad_norm": 1.2363786697387695,
|
1256 |
+
"learning_rate": 1.0217391304347827e-05,
|
1257 |
+
"loss": 0.6239,
|
1258 |
+
"step": 70000
|
1259 |
+
},
|
1260 |
+
{
|
1261 |
+
"epoch": 4.0,
|
1262 |
+
"eval_loss": 1.0999037027359009,
|
1263 |
+
"eval_runtime": 239.6686,
|
1264 |
+
"eval_samples_per_second": 251.493,
|
1265 |
+
"eval_steps_per_second": 3.93,
|
1266 |
+
"step": 70192
|
1267 |
+
},
|
1268 |
+
{
|
1269 |
+
"epoch": 4.01,
|
1270 |
+
"grad_norm": 1.0348886251449585,
|
1271 |
+
"learning_rate": 9.98734759604325e-06,
|
1272 |
+
"loss": 0.6073,
|
1273 |
+
"step": 70400
|
1274 |
+
},
|
1275 |
+
{
|
1276 |
+
"epoch": 4.03,
|
1277 |
+
"grad_norm": 1.2015148401260376,
|
1278 |
+
"learning_rate": 9.757303887738671e-06,
|
1279 |
+
"loss": 0.5995,
|
1280 |
+
"step": 70800
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 4.06,
|
1284 |
+
"grad_norm": 1.2094097137451172,
|
1285 |
+
"learning_rate": 9.527260179434092e-06,
|
1286 |
+
"loss": 0.5939,
|
1287 |
+
"step": 71200
|
1288 |
+
},
|
1289 |
+
{
|
1290 |
+
"epoch": 4.08,
|
1291 |
+
"grad_norm": 1.1249605417251587,
|
1292 |
+
"learning_rate": 9.297216471129515e-06,
|
1293 |
+
"loss": 0.6087,
|
1294 |
+
"step": 71600
|
1295 |
+
},
|
1296 |
+
{
|
1297 |
+
"epoch": 4.1,
|
1298 |
+
"grad_norm": 1.1234875917434692,
|
1299 |
+
"learning_rate": 9.067172762824937e-06,
|
1300 |
+
"loss": 0.5966,
|
1301 |
+
"step": 72000
|
1302 |
+
},
|
1303 |
+
{
|
1304 |
+
"epoch": 4.13,
|
1305 |
+
"grad_norm": 1.3029675483703613,
|
1306 |
+
"learning_rate": 8.837704163791121e-06,
|
1307 |
+
"loss": 0.6023,
|
1308 |
+
"step": 72400
|
1309 |
+
},
|
1310 |
+
{
|
1311 |
+
"epoch": 4.15,
|
1312 |
+
"grad_norm": 1.1492359638214111,
|
1313 |
+
"learning_rate": 8.607660455486542e-06,
|
1314 |
+
"loss": 0.6014,
|
1315 |
+
"step": 72800
|
1316 |
+
},
|
1317 |
+
{
|
1318 |
+
"epoch": 4.17,
|
1319 |
+
"grad_norm": 1.1293883323669434,
|
1320 |
+
"learning_rate": 8.377616747181965e-06,
|
1321 |
+
"loss": 0.5975,
|
1322 |
+
"step": 73200
|
1323 |
+
},
|
1324 |
+
{
|
1325 |
+
"epoch": 4.19,
|
1326 |
+
"grad_norm": 1.1434813737869263,
|
1327 |
+
"learning_rate": 8.147573038877387e-06,
|
1328 |
+
"loss": 0.5928,
|
1329 |
+
"step": 73600
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 4.22,
|
1333 |
+
"grad_norm": 1.2983373403549194,
|
1334 |
+
"learning_rate": 7.91752933057281e-06,
|
1335 |
+
"loss": 0.5987,
|
1336 |
+
"step": 74000
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 4.24,
|
1340 |
+
"grad_norm": 1.190185785293579,
|
1341 |
+
"learning_rate": 7.687485622268231e-06,
|
1342 |
+
"loss": 0.5948,
|
1343 |
+
"step": 74400
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 4.26,
|
1347 |
+
"grad_norm": 1.068985939025879,
|
1348 |
+
"learning_rate": 7.457441913963653e-06,
|
1349 |
+
"loss": 0.6059,
|
1350 |
+
"step": 74800
|
1351 |
+
},
|
1352 |
+
{
|
1353 |
+
"epoch": 4.29,
|
1354 |
+
"grad_norm": 1.0893745422363281,
|
1355 |
+
"learning_rate": 7.227398205659075e-06,
|
1356 |
+
"loss": 0.61,
|
1357 |
+
"step": 75200
|
1358 |
+
},
|
1359 |
+
{
|
1360 |
+
"epoch": 4.31,
|
1361 |
+
"grad_norm": 1.1232637166976929,
|
1362 |
+
"learning_rate": 6.997354497354498e-06,
|
1363 |
+
"loss": 0.5947,
|
1364 |
+
"step": 75600
|
1365 |
+
},
|
1366 |
+
{
|
1367 |
+
"epoch": 4.33,
|
1368 |
+
"grad_norm": 1.1357260942459106,
|
1369 |
+
"learning_rate": 6.76731078904992e-06,
|
1370 |
+
"loss": 0.596,
|
1371 |
+
"step": 76000
|
1372 |
+
},
|
1373 |
+
{
|
1374 |
+
"epoch": 4.35,
|
1375 |
+
"grad_norm": 1.1744706630706787,
|
1376 |
+
"learning_rate": 6.537267080745342e-06,
|
1377 |
+
"loss": 0.5909,
|
1378 |
+
"step": 76400
|
1379 |
+
},
|
1380 |
+
{
|
1381 |
+
"epoch": 4.38,
|
1382 |
+
"grad_norm": 1.2433269023895264,
|
1383 |
+
"learning_rate": 6.307223372440764e-06,
|
1384 |
+
"loss": 0.6008,
|
1385 |
+
"step": 76800
|
1386 |
+
},
|
1387 |
+
{
|
1388 |
+
"epoch": 4.4,
|
1389 |
+
"grad_norm": 1.1302207708358765,
|
1390 |
+
"learning_rate": 6.0777547734069475e-06,
|
1391 |
+
"loss": 0.6091,
|
1392 |
+
"step": 77200
|
1393 |
+
},
|
1394 |
+
{
|
1395 |
+
"epoch": 4.42,
|
1396 |
+
"grad_norm": 1.309975504875183,
|
1397 |
+
"learning_rate": 5.848286174373131e-06,
|
1398 |
+
"loss": 0.5954,
|
1399 |
+
"step": 77600
|
1400 |
+
},
|
1401 |
+
{
|
1402 |
+
"epoch": 4.44,
|
1403 |
+
"grad_norm": 1.2409945726394653,
|
1404 |
+
"learning_rate": 5.618242466068553e-06,
|
1405 |
+
"loss": 0.5986,
|
1406 |
+
"step": 78000
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 4.47,
|
1410 |
+
"grad_norm": 1.274553894996643,
|
1411 |
+
"learning_rate": 5.388198757763975e-06,
|
1412 |
+
"loss": 0.6031,
|
1413 |
+
"step": 78400
|
1414 |
+
},
|
1415 |
+
{
|
1416 |
+
"epoch": 4.49,
|
1417 |
+
"grad_norm": 1.2017253637313843,
|
1418 |
+
"learning_rate": 5.1581550494593975e-06,
|
1419 |
+
"loss": 0.5944,
|
1420 |
+
"step": 78800
|
1421 |
+
},
|
1422 |
+
{
|
1423 |
+
"epoch": 4.51,
|
1424 |
+
"grad_norm": 1.2163805961608887,
|
1425 |
+
"learning_rate": 4.92811134115482e-06,
|
1426 |
+
"loss": 0.5921,
|
1427 |
+
"step": 79200
|
1428 |
+
},
|
1429 |
+
{
|
1430 |
+
"epoch": 4.54,
|
1431 |
+
"grad_norm": 1.2417266368865967,
|
1432 |
+
"learning_rate": 4.698067632850242e-06,
|
1433 |
+
"loss": 0.5897,
|
1434 |
+
"step": 79600
|
1435 |
+
},
|
1436 |
+
{
|
1437 |
+
"epoch": 4.56,
|
1438 |
+
"grad_norm": 1.212352991104126,
|
1439 |
+
"learning_rate": 4.468023924545664e-06,
|
1440 |
+
"loss": 0.6028,
|
1441 |
+
"step": 80000
|
1442 |
+
},
|
1443 |
+
{
|
1444 |
+
"epoch": 4.58,
|
1445 |
+
"grad_norm": 1.2147575616836548,
|
1446 |
+
"learning_rate": 4.237980216241086e-06,
|
1447 |
+
"loss": 0.5964,
|
1448 |
+
"step": 80400
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 4.6,
|
1452 |
+
"grad_norm": 1.315047025680542,
|
1453 |
+
"learning_rate": 4.007936507936508e-06,
|
1454 |
+
"loss": 0.6008,
|
1455 |
+
"step": 80800
|
1456 |
+
},
|
1457 |
+
{
|
1458 |
+
"epoch": 4.63,
|
1459 |
+
"grad_norm": 1.3426513671875,
|
1460 |
+
"learning_rate": 3.7778927996319303e-06,
|
1461 |
+
"loss": 0.5913,
|
1462 |
+
"step": 81200
|
1463 |
+
},
|
1464 |
+
{
|
1465 |
+
"epoch": 4.65,
|
1466 |
+
"grad_norm": 1.1529968976974487,
|
1467 |
+
"learning_rate": 3.5478490913273524e-06,
|
1468 |
+
"loss": 0.5966,
|
1469 |
+
"step": 81600
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 4.67,
|
1473 |
+
"grad_norm": 1.2418185472488403,
|
1474 |
+
"learning_rate": 3.317805383022775e-06,
|
1475 |
+
"loss": 0.5878,
|
1476 |
+
"step": 82000
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 4.7,
|
1480 |
+
"grad_norm": 1.1235523223876953,
|
1481 |
+
"learning_rate": 3.0877616747181967e-06,
|
1482 |
+
"loss": 0.5931,
|
1483 |
+
"step": 82400
|
1484 |
+
},
|
1485 |
+
{
|
1486 |
+
"epoch": 4.72,
|
1487 |
+
"grad_norm": 1.3443814516067505,
|
1488 |
+
"learning_rate": 2.85829307568438e-06,
|
1489 |
+
"loss": 0.6064,
|
1490 |
+
"step": 82800
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 4.74,
|
1494 |
+
"grad_norm": 1.2628613710403442,
|
1495 |
+
"learning_rate": 2.6288244766505636e-06,
|
1496 |
+
"loss": 0.6001,
|
1497 |
+
"step": 83200
|
1498 |
+
},
|
1499 |
+
{
|
1500 |
+
"epoch": 4.76,
|
1501 |
+
"grad_norm": 1.1324889659881592,
|
1502 |
+
"learning_rate": 2.3987807683459858e-06,
|
1503 |
+
"loss": 0.5916,
|
1504 |
+
"step": 83600
|
1505 |
+
},
|
1506 |
+
{
|
1507 |
+
"epoch": 4.79,
|
1508 |
+
"grad_norm": 1.1024839878082275,
|
1509 |
+
"learning_rate": 2.168737060041408e-06,
|
1510 |
+
"loss": 0.6048,
|
1511 |
+
"step": 84000
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 4.81,
|
1515 |
+
"grad_norm": 1.3531837463378906,
|
1516 |
+
"learning_rate": 1.93869335173683e-06,
|
1517 |
+
"loss": 0.5896,
|
1518 |
+
"step": 84400
|
1519 |
+
},
|
1520 |
+
{
|
1521 |
+
"epoch": 4.83,
|
1522 |
+
"grad_norm": 1.1726586818695068,
|
1523 |
+
"learning_rate": 1.7086496434322524e-06,
|
1524 |
+
"loss": 0.5905,
|
1525 |
+
"step": 84800
|
1526 |
+
},
|
1527 |
+
{
|
1528 |
+
"epoch": 4.86,
|
1529 |
+
"grad_norm": 1.0651309490203857,
|
1530 |
+
"learning_rate": 1.4791810443984358e-06,
|
1531 |
+
"loss": 0.6002,
|
1532 |
+
"step": 85200
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 4.88,
|
1536 |
+
"grad_norm": 1.277769684791565,
|
1537 |
+
"learning_rate": 1.2497124453646193e-06,
|
1538 |
+
"loss": 0.5899,
|
1539 |
+
"step": 85600
|
1540 |
+
},
|
1541 |
+
{
|
1542 |
+
"epoch": 4.9,
|
1543 |
+
"grad_norm": 1.2923473119735718,
|
1544 |
+
"learning_rate": 1.0196687370600414e-06,
|
1545 |
+
"loss": 0.5924,
|
1546 |
+
"step": 86000
|
1547 |
+
},
|
1548 |
+
{
|
1549 |
+
"epoch": 4.92,
|
1550 |
+
"grad_norm": 1.1012171506881714,
|
1551 |
+
"learning_rate": 7.896250287554636e-07,
|
1552 |
+
"loss": 0.591,
|
1553 |
+
"step": 86400
|
1554 |
+
},
|
1555 |
+
{
|
1556 |
+
"epoch": 4.95,
|
1557 |
+
"grad_norm": 1.2737475633621216,
|
1558 |
+
"learning_rate": 5.595813204508857e-07,
|
1559 |
+
"loss": 0.5968,
|
1560 |
+
"step": 86800
|
1561 |
+
},
|
1562 |
+
{
|
1563 |
+
"epoch": 4.97,
|
1564 |
+
"grad_norm": 1.2628225088119507,
|
1565 |
+
"learning_rate": 3.295376121463078e-07,
|
1566 |
+
"loss": 0.591,
|
1567 |
+
"step": 87200
|
1568 |
+
},
|
1569 |
+
{
|
1570 |
+
"epoch": 4.99,
|
1571 |
+
"grad_norm": 1.3517860174179077,
|
1572 |
+
"learning_rate": 9.949390384172993e-08,
|
1573 |
+
"loss": 0.5918,
|
1574 |
+
"step": 87600
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 5.0,
|
1578 |
+
"eval_loss": 1.1124831438064575,
|
1579 |
+
"eval_runtime": 239.5934,
|
1580 |
+
"eval_samples_per_second": 251.572,
|
1581 |
+
"eval_steps_per_second": 3.932,
|
1582 |
+
"step": 87740
|
1583 |
+
},
|
1584 |
+
{
|
1585 |
+
"epoch": 5.0,
|
1586 |
+
"step": 87740,
|
1587 |
+
"total_flos": 1.7097588901675008e+18,
|
1588 |
+
"train_loss": 0.74035006675955,
|
1589 |
+
"train_runtime": 31417.8788,
|
1590 |
+
"train_samples_per_second": 89.369,
|
1591 |
+
"train_steps_per_second": 2.793
|
1592 |
+
}
|
1593 |
+
],
|
1594 |
+
"logging_steps": 400,
|
1595 |
+
"max_steps": 87740,
|
1596 |
+
"num_input_tokens_seen": 0,
|
1597 |
+
"num_train_epochs": 5,
|
1598 |
+
"save_steps": 500,
|
1599 |
+
"total_flos": 1.7097588901675008e+18,
|
1600 |
+
"train_batch_size": 32,
|
1601 |
+
"trial_name": null,
|
1602 |
+
"trial_params": null
|
1603 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:061977bcd4abed9cc949a4acc04dd073a7a4857180d7ce144d844b2e9091cb9c
|
3 |
+
size 5176
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|