theblackcat102
commited on
Commit
•
24b78a2
1
Parent(s):
2d05120
Upload 13 files
Browse files- added_tokens.json +3 -0
- config.json +42 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +9 -0
- spm.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +17 -0
- trainer_state.json +1786 -0
- training_args.bin +3 -0
added_tokens.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"[MASK]": 128000
|
3 |
+
}
|
config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "microsoft/deberta-v3-large",
|
3 |
+
"architectures": [
|
4 |
+
"DebertaV2ForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"hidden_act": "gelu",
|
8 |
+
"hidden_dropout_prob": 0.1,
|
9 |
+
"hidden_size": 1024,
|
10 |
+
"id2label": {
|
11 |
+
"0": "REWARD"
|
12 |
+
},
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 4096,
|
15 |
+
"label2id": {
|
16 |
+
"REWARD": 0
|
17 |
+
},
|
18 |
+
"layer_norm_eps": 1e-07,
|
19 |
+
"max_position_embeddings": 512,
|
20 |
+
"max_relative_positions": -1,
|
21 |
+
"model_type": "deberta-v2",
|
22 |
+
"norm_rel_ebd": "layer_norm",
|
23 |
+
"num_attention_heads": 16,
|
24 |
+
"num_hidden_layers": 24,
|
25 |
+
"pad_token_id": 0,
|
26 |
+
"pooler_dropout": 0,
|
27 |
+
"pooler_hidden_act": "gelu",
|
28 |
+
"pooler_hidden_size": 1024,
|
29 |
+
"pos_att_type": [
|
30 |
+
"p2c",
|
31 |
+
"c2p"
|
32 |
+
],
|
33 |
+
"position_biased_input": false,
|
34 |
+
"position_buckets": 256,
|
35 |
+
"problem_type": "regression",
|
36 |
+
"relative_attention": true,
|
37 |
+
"share_att_key": true,
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.25.1",
|
40 |
+
"type_vocab_size": 0,
|
41 |
+
"vocab_size": 128100
|
42 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9ae362c80bd91e9e73c9aac411565ddadb46ac62ac86d0790b0a5b0319472fef
|
3 |
+
size 3480731481
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:903103d4990515d537c46262632aef9a8f7d0a815edd289868c9967e5b15c265
|
3 |
+
size 1740389291
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c2428b8cac3bfdf85570df6764e2e1f50da404b41712682d4c39309cd3074ea2
|
3 |
+
size 14567
|
scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65725e6f74c1a76777f848d1c32499753466866e08e36dea01cdb5d6bbb40c1a
|
3 |
+
size 559
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:36d48408117e3b27e140662f5d18621485e88b5d1353555e6bcc7c8ee1a9de88
|
3 |
+
size 559
|
special_tokens_map.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "[CLS]",
|
3 |
+
"cls_token": "[CLS]",
|
4 |
+
"eos_token": "[SEP]",
|
5 |
+
"mask_token": "[MASK]",
|
6 |
+
"pad_token": "[PAD]",
|
7 |
+
"sep_token": "[SEP]",
|
8 |
+
"unk_token": "[UNK]"
|
9 |
+
}
|
spm.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
3 |
+
size 2464616
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "[CLS]",
|
3 |
+
"cls_token": "[CLS]",
|
4 |
+
"do_lower_case": false,
|
5 |
+
"eos_token": "[SEP]",
|
6 |
+
"mask_token": "[MASK]",
|
7 |
+
"model_max_length": 1000000000000000019884624838656,
|
8 |
+
"name_or_path": "microsoft/deberta-v3-large",
|
9 |
+
"pad_token": "[PAD]",
|
10 |
+
"sep_token": "[SEP]",
|
11 |
+
"sp_model_kwargs": {},
|
12 |
+
"special_tokens_map_file": null,
|
13 |
+
"split_by_punct": false,
|
14 |
+
"tokenizer_class": "DebertaV2Tokenizer",
|
15 |
+
"unk_token": "[UNK]",
|
16 |
+
"vocab_type": "spm"
|
17 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1786 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.301327088212334,
|
5 |
+
"global_step": 2500,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 1.8e-07,
|
13 |
+
"loss": 0.6935,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.01,
|
18 |
+
"learning_rate": 3.8e-07,
|
19 |
+
"loss": 0.6974,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.02,
|
24 |
+
"learning_rate": 5.800000000000001e-07,
|
25 |
+
"loss": 0.694,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.02,
|
30 |
+
"learning_rate": 7.8e-07,
|
31 |
+
"loss": 0.6966,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.03,
|
36 |
+
"learning_rate": 9.800000000000001e-07,
|
37 |
+
"loss": 0.6948,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.03,
|
42 |
+
"learning_rate": 1.1800000000000001e-06,
|
43 |
+
"loss": 0.6957,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.04,
|
48 |
+
"learning_rate": 1.3800000000000001e-06,
|
49 |
+
"loss": 0.6952,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.04,
|
54 |
+
"learning_rate": 1.5800000000000001e-06,
|
55 |
+
"loss": 0.6972,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.05,
|
60 |
+
"learning_rate": 1.7800000000000001e-06,
|
61 |
+
"loss": 0.6925,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.05,
|
66 |
+
"learning_rate": 1.98e-06,
|
67 |
+
"loss": 0.6903,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.06,
|
72 |
+
"learning_rate": 2.1800000000000003e-06,
|
73 |
+
"loss": 0.6837,
|
74 |
+
"step": 110
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.06,
|
78 |
+
"learning_rate": 2.38e-06,
|
79 |
+
"loss": 0.6754,
|
80 |
+
"step": 120
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.07,
|
84 |
+
"learning_rate": 2.5800000000000003e-06,
|
85 |
+
"loss": 0.6447,
|
86 |
+
"step": 130
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.07,
|
90 |
+
"learning_rate": 2.7800000000000005e-06,
|
91 |
+
"loss": 0.6391,
|
92 |
+
"step": 140
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.08,
|
96 |
+
"learning_rate": 2.9800000000000003e-06,
|
97 |
+
"loss": 0.6118,
|
98 |
+
"step": 150
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.08,
|
102 |
+
"learning_rate": 3.1800000000000005e-06,
|
103 |
+
"loss": 0.6071,
|
104 |
+
"step": 160
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.09,
|
108 |
+
"learning_rate": 3.3800000000000007e-06,
|
109 |
+
"loss": 0.6027,
|
110 |
+
"step": 170
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.09,
|
114 |
+
"learning_rate": 3.58e-06,
|
115 |
+
"loss": 0.5978,
|
116 |
+
"step": 180
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.1,
|
120 |
+
"learning_rate": 3.7800000000000002e-06,
|
121 |
+
"loss": 0.5974,
|
122 |
+
"step": 190
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.1,
|
126 |
+
"learning_rate": 3.980000000000001e-06,
|
127 |
+
"loss": 0.5929,
|
128 |
+
"step": 200
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.11,
|
132 |
+
"learning_rate": 4.18e-06,
|
133 |
+
"loss": 0.5853,
|
134 |
+
"step": 210
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.11,
|
138 |
+
"learning_rate": 4.38e-06,
|
139 |
+
"loss": 0.582,
|
140 |
+
"step": 220
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.12,
|
144 |
+
"learning_rate": 4.58e-06,
|
145 |
+
"loss": 0.5443,
|
146 |
+
"step": 230
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.12,
|
150 |
+
"learning_rate": 4.78e-06,
|
151 |
+
"loss": 0.5475,
|
152 |
+
"step": 240
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.13,
|
156 |
+
"learning_rate": 4.980000000000001e-06,
|
157 |
+
"loss": 0.5348,
|
158 |
+
"step": 250
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.13,
|
162 |
+
"eval_webgpt_accuracy": 0.5572012257405515,
|
163 |
+
"eval_webgpt_loss": 0.6882389783859253,
|
164 |
+
"eval_webgpt_runtime": 166.4816,
|
165 |
+
"eval_webgpt_samples_per_second": 23.522,
|
166 |
+
"eval_webgpt_steps_per_second": 4.709,
|
167 |
+
"step": 250
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 0.13,
|
171 |
+
"eval_hfsummary_accuracy": 0.6711301877096999,
|
172 |
+
"eval_hfsummary_loss": 0.6183204054832458,
|
173 |
+
"eval_hfsummary_runtime": 3223.0913,
|
174 |
+
"eval_hfsummary_samples_per_second": 10.264,
|
175 |
+
"eval_hfsummary_steps_per_second": 2.053,
|
176 |
+
"step": 250
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.13,
|
180 |
+
"eval_gptsynthetic_accuracy": 0.9993966817496229,
|
181 |
+
"eval_gptsynthetic_loss": 0.009097465313971043,
|
182 |
+
"eval_gptsynthetic_runtime": 101.5011,
|
183 |
+
"eval_gptsynthetic_samples_per_second": 32.66,
|
184 |
+
"eval_gptsynthetic_steps_per_second": 6.532,
|
185 |
+
"step": 250
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 0.14,
|
189 |
+
"learning_rate": 5.18e-06,
|
190 |
+
"loss": 0.537,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.14,
|
195 |
+
"learning_rate": 5.380000000000001e-06,
|
196 |
+
"loss": 0.5286,
|
197 |
+
"step": 270
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 0.15,
|
201 |
+
"learning_rate": 5.580000000000001e-06,
|
202 |
+
"loss": 0.4957,
|
203 |
+
"step": 280
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 0.15,
|
207 |
+
"learning_rate": 5.78e-06,
|
208 |
+
"loss": 0.5117,
|
209 |
+
"step": 290
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.16,
|
213 |
+
"learning_rate": 5.98e-06,
|
214 |
+
"loss": 0.4977,
|
215 |
+
"step": 300
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 0.16,
|
219 |
+
"learning_rate": 6.18e-06,
|
220 |
+
"loss": 0.4801,
|
221 |
+
"step": 310
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.17,
|
225 |
+
"learning_rate": 6.380000000000001e-06,
|
226 |
+
"loss": 0.499,
|
227 |
+
"step": 320
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.17,
|
231 |
+
"learning_rate": 6.5800000000000005e-06,
|
232 |
+
"loss": 0.4702,
|
233 |
+
"step": 330
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.18,
|
237 |
+
"learning_rate": 6.780000000000001e-06,
|
238 |
+
"loss": 0.4715,
|
239 |
+
"step": 340
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.18,
|
243 |
+
"learning_rate": 6.98e-06,
|
244 |
+
"loss": 0.4686,
|
245 |
+
"step": 350
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 0.19,
|
249 |
+
"learning_rate": 7.180000000000001e-06,
|
250 |
+
"loss": 0.488,
|
251 |
+
"step": 360
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 0.19,
|
255 |
+
"learning_rate": 7.3800000000000005e-06,
|
256 |
+
"loss": 0.4671,
|
257 |
+
"step": 370
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.2,
|
261 |
+
"learning_rate": 7.58e-06,
|
262 |
+
"loss": 0.4889,
|
263 |
+
"step": 380
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 0.2,
|
267 |
+
"learning_rate": 7.78e-06,
|
268 |
+
"loss": 0.4718,
|
269 |
+
"step": 390
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.21,
|
273 |
+
"learning_rate": 7.980000000000002e-06,
|
274 |
+
"loss": 0.4706,
|
275 |
+
"step": 400
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.21,
|
279 |
+
"learning_rate": 8.18e-06,
|
280 |
+
"loss": 0.4839,
|
281 |
+
"step": 410
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 0.22,
|
285 |
+
"learning_rate": 8.380000000000001e-06,
|
286 |
+
"loss": 0.4576,
|
287 |
+
"step": 420
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 0.22,
|
291 |
+
"learning_rate": 8.580000000000001e-06,
|
292 |
+
"loss": 0.4865,
|
293 |
+
"step": 430
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 0.23,
|
297 |
+
"learning_rate": 8.78e-06,
|
298 |
+
"loss": 0.468,
|
299 |
+
"step": 440
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 0.23,
|
303 |
+
"learning_rate": 8.98e-06,
|
304 |
+
"loss": 0.4335,
|
305 |
+
"step": 450
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 0.24,
|
309 |
+
"learning_rate": 9.180000000000002e-06,
|
310 |
+
"loss": 0.4516,
|
311 |
+
"step": 460
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 0.24,
|
315 |
+
"learning_rate": 9.38e-06,
|
316 |
+
"loss": 0.4603,
|
317 |
+
"step": 470
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.25,
|
321 |
+
"learning_rate": 9.58e-06,
|
322 |
+
"loss": 0.467,
|
323 |
+
"step": 480
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 0.26,
|
327 |
+
"learning_rate": 9.780000000000001e-06,
|
328 |
+
"loss": 0.4678,
|
329 |
+
"step": 490
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 0.26,
|
333 |
+
"learning_rate": 9.980000000000001e-06,
|
334 |
+
"loss": 0.4456,
|
335 |
+
"step": 500
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 0.26,
|
339 |
+
"eval_webgpt_accuracy": 0.5893769152196119,
|
340 |
+
"eval_webgpt_loss": 0.6772251725196838,
|
341 |
+
"eval_webgpt_runtime": 166.9484,
|
342 |
+
"eval_webgpt_samples_per_second": 23.456,
|
343 |
+
"eval_webgpt_steps_per_second": 4.696,
|
344 |
+
"step": 500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.26,
|
348 |
+
"eval_hfsummary_accuracy": 0.7065864643472478,
|
349 |
+
"eval_hfsummary_loss": 0.5687452554702759,
|
350 |
+
"eval_hfsummary_runtime": 3231.8075,
|
351 |
+
"eval_hfsummary_samples_per_second": 10.237,
|
352 |
+
"eval_hfsummary_steps_per_second": 2.047,
|
353 |
+
"step": 500
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 0.26,
|
357 |
+
"eval_gptsynthetic_accuracy": 0.9993966817496229,
|
358 |
+
"eval_gptsynthetic_loss": 0.005173402838408947,
|
359 |
+
"eval_gptsynthetic_runtime": 101.6573,
|
360 |
+
"eval_gptsynthetic_samples_per_second": 32.61,
|
361 |
+
"eval_gptsynthetic_steps_per_second": 6.522,
|
362 |
+
"step": 500
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 0.27,
|
366 |
+
"learning_rate": 9.999821166271659e-06,
|
367 |
+
"loss": 0.4375,
|
368 |
+
"step": 510
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 0.27,
|
372 |
+
"learning_rate": 9.999202992029537e-06,
|
373 |
+
"loss": 0.4608,
|
374 |
+
"step": 520
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.28,
|
378 |
+
"learning_rate": 9.998143324029546e-06,
|
379 |
+
"loss": 0.4674,
|
380 |
+
"step": 530
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.28,
|
384 |
+
"learning_rate": 9.996642255853966e-06,
|
385 |
+
"loss": 0.4468,
|
386 |
+
"step": 540
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 0.29,
|
390 |
+
"learning_rate": 9.99469992006637e-06,
|
391 |
+
"loss": 0.4466,
|
392 |
+
"step": 550
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 0.29,
|
396 |
+
"learning_rate": 9.992316488199927e-06,
|
397 |
+
"loss": 0.4608,
|
398 |
+
"step": 560
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 0.3,
|
402 |
+
"learning_rate": 9.989492170742248e-06,
|
403 |
+
"loss": 0.436,
|
404 |
+
"step": 570
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 0.3,
|
408 |
+
"learning_rate": 9.986227217116798e-06,
|
409 |
+
"loss": 0.4249,
|
410 |
+
"step": 580
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 0.31,
|
414 |
+
"learning_rate": 9.982521915660865e-06,
|
415 |
+
"loss": 0.4702,
|
416 |
+
"step": 590
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.31,
|
420 |
+
"learning_rate": 9.978376593600107e-06,
|
421 |
+
"loss": 0.4336,
|
422 |
+
"step": 600
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.32,
|
426 |
+
"learning_rate": 9.973791617019634e-06,
|
427 |
+
"loss": 0.4617,
|
428 |
+
"step": 610
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.32,
|
432 |
+
"learning_rate": 9.968767390831707e-06,
|
433 |
+
"loss": 0.4547,
|
434 |
+
"step": 620
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.33,
|
438 |
+
"learning_rate": 9.963304358739943e-06,
|
439 |
+
"loss": 0.4285,
|
440 |
+
"step": 630
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.33,
|
444 |
+
"learning_rate": 9.957403003200167e-06,
|
445 |
+
"loss": 0.4626,
|
446 |
+
"step": 640
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.34,
|
450 |
+
"learning_rate": 9.951063845377778e-06,
|
451 |
+
"loss": 0.4153,
|
452 |
+
"step": 650
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.34,
|
456 |
+
"learning_rate": 9.944287445101733e-06,
|
457 |
+
"loss": 0.4196,
|
458 |
+
"step": 660
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.35,
|
462 |
+
"learning_rate": 9.937074400815114e-06,
|
463 |
+
"loss": 0.4461,
|
464 |
+
"step": 670
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.35,
|
468 |
+
"learning_rate": 9.929425349522264e-06,
|
469 |
+
"loss": 0.4108,
|
470 |
+
"step": 680
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.36,
|
474 |
+
"learning_rate": 9.921340966732539e-06,
|
475 |
+
"loss": 0.4466,
|
476 |
+
"step": 690
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.36,
|
480 |
+
"learning_rate": 9.912821966400652e-06,
|
481 |
+
"loss": 0.4608,
|
482 |
+
"step": 700
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 0.37,
|
486 |
+
"learning_rate": 9.903869100863622e-06,
|
487 |
+
"loss": 0.4382,
|
488 |
+
"step": 710
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 0.37,
|
492 |
+
"learning_rate": 9.894483160774331e-06,
|
493 |
+
"loss": 0.4172,
|
494 |
+
"step": 720
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 0.38,
|
498 |
+
"learning_rate": 9.884664975031698e-06,
|
499 |
+
"loss": 0.4021,
|
500 |
+
"step": 730
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 0.39,
|
504 |
+
"learning_rate": 9.874415410707476e-06,
|
505 |
+
"loss": 0.4108,
|
506 |
+
"step": 740
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.39,
|
510 |
+
"learning_rate": 9.863735372969692e-06,
|
511 |
+
"loss": 0.4402,
|
512 |
+
"step": 750
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 0.39,
|
516 |
+
"eval_webgpt_accuracy": 0.5947395301327886,
|
517 |
+
"eval_webgpt_loss": 0.6755640506744385,
|
518 |
+
"eval_webgpt_runtime": 167.2694,
|
519 |
+
"eval_webgpt_samples_per_second": 23.411,
|
520 |
+
"eval_webgpt_steps_per_second": 4.687,
|
521 |
+
"step": 750
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 0.39,
|
525 |
+
"eval_hfsummary_accuracy": 0.7158661548227186,
|
526 |
+
"eval_hfsummary_loss": 0.5612313151359558,
|
527 |
+
"eval_hfsummary_runtime": 3314.4907,
|
528 |
+
"eval_hfsummary_samples_per_second": 9.981,
|
529 |
+
"eval_hfsummary_steps_per_second": 1.996,
|
530 |
+
"step": 750
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 0.39,
|
534 |
+
"eval_gptsynthetic_accuracy": 0.9996983408748115,
|
535 |
+
"eval_gptsynthetic_loss": 0.005162315908819437,
|
536 |
+
"eval_gptsynthetic_runtime": 105.2324,
|
537 |
+
"eval_gptsynthetic_samples_per_second": 31.502,
|
538 |
+
"eval_gptsynthetic_steps_per_second": 6.3,
|
539 |
+
"step": 750
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 0.4,
|
543 |
+
"learning_rate": 9.852625805002682e-06,
|
544 |
+
"loss": 0.4411,
|
545 |
+
"step": 760
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 0.4,
|
549 |
+
"learning_rate": 9.841087687923826e-06,
|
550 |
+
"loss": 0.4239,
|
551 |
+
"step": 770
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 0.41,
|
555 |
+
"learning_rate": 9.82912204069688e-06,
|
556 |
+
"loss": 0.4489,
|
557 |
+
"step": 780
|
558 |
+
},
|
559 |
+
{
|
560 |
+
"epoch": 0.41,
|
561 |
+
"learning_rate": 9.816729920041996e-06,
|
562 |
+
"loss": 0.4337,
|
563 |
+
"step": 790
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 0.42,
|
567 |
+
"learning_rate": 9.803912420342406e-06,
|
568 |
+
"loss": 0.4346,
|
569 |
+
"step": 800
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.42,
|
573 |
+
"learning_rate": 9.790670673547766e-06,
|
574 |
+
"loss": 0.3994,
|
575 |
+
"step": 810
|
576 |
+
},
|
577 |
+
{
|
578 |
+
"epoch": 0.43,
|
579 |
+
"learning_rate": 9.777005849074195e-06,
|
580 |
+
"loss": 0.4151,
|
581 |
+
"step": 820
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 0.43,
|
585 |
+
"learning_rate": 9.762919153700996e-06,
|
586 |
+
"loss": 0.4519,
|
587 |
+
"step": 830
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 0.44,
|
591 |
+
"learning_rate": 9.748411831464083e-06,
|
592 |
+
"loss": 0.4409,
|
593 |
+
"step": 840
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 0.44,
|
597 |
+
"learning_rate": 9.733485163546124e-06,
|
598 |
+
"loss": 0.4069,
|
599 |
+
"step": 850
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 0.45,
|
603 |
+
"learning_rate": 9.718140468163386e-06,
|
604 |
+
"loss": 0.3974,
|
605 |
+
"step": 860
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 0.45,
|
609 |
+
"learning_rate": 9.702379100449325e-06,
|
610 |
+
"loss": 0.4192,
|
611 |
+
"step": 870
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.46,
|
615 |
+
"learning_rate": 9.686202452334909e-06,
|
616 |
+
"loss": 0.4106,
|
617 |
+
"step": 880
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 0.46,
|
621 |
+
"learning_rate": 9.669611952425694e-06,
|
622 |
+
"loss": 0.4224,
|
623 |
+
"step": 890
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 0.47,
|
627 |
+
"learning_rate": 9.652609065875656e-06,
|
628 |
+
"loss": 0.4176,
|
629 |
+
"step": 900
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 0.47,
|
633 |
+
"learning_rate": 9.635195294257806e-06,
|
634 |
+
"loss": 0.3897,
|
635 |
+
"step": 910
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 0.48,
|
639 |
+
"learning_rate": 9.617372175431574e-06,
|
640 |
+
"loss": 0.4201,
|
641 |
+
"step": 920
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 0.48,
|
645 |
+
"learning_rate": 9.599141283407004e-06,
|
646 |
+
"loss": 0.3955,
|
647 |
+
"step": 930
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 0.49,
|
651 |
+
"learning_rate": 9.580504228205736e-06,
|
652 |
+
"loss": 0.3754,
|
653 |
+
"step": 940
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.49,
|
657 |
+
"learning_rate": 9.56146265571884e-06,
|
658 |
+
"loss": 0.407,
|
659 |
+
"step": 950
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 0.5,
|
663 |
+
"learning_rate": 9.542018247561442e-06,
|
664 |
+
"loss": 0.4225,
|
665 |
+
"step": 960
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 0.5,
|
669 |
+
"learning_rate": 9.522172720924236e-06,
|
670 |
+
"loss": 0.415,
|
671 |
+
"step": 970
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 0.51,
|
675 |
+
"learning_rate": 9.501927828421816e-06,
|
676 |
+
"loss": 0.4014,
|
677 |
+
"step": 980
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 0.52,
|
681 |
+
"learning_rate": 9.481285357937905e-06,
|
682 |
+
"loss": 0.4074,
|
683 |
+
"step": 990
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 0.52,
|
687 |
+
"learning_rate": 9.460247132467466e-06,
|
688 |
+
"loss": 0.3874,
|
689 |
+
"step": 1000
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.52,
|
693 |
+
"eval_webgpt_accuracy": 0.6021450459652706,
|
694 |
+
"eval_webgpt_loss": 0.6651596426963806,
|
695 |
+
"eval_webgpt_runtime": 167.2303,
|
696 |
+
"eval_webgpt_samples_per_second": 23.417,
|
697 |
+
"eval_webgpt_steps_per_second": 4.688,
|
698 |
+
"step": 1000
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 0.52,
|
702 |
+
"eval_hfsummary_accuracy": 0.714052534534353,
|
703 |
+
"eval_hfsummary_loss": 0.5728897452354431,
|
704 |
+
"eval_hfsummary_runtime": 3240.5714,
|
705 |
+
"eval_hfsummary_samples_per_second": 10.209,
|
706 |
+
"eval_hfsummary_steps_per_second": 2.042,
|
707 |
+
"step": 1000
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 0.52,
|
711 |
+
"eval_gptsynthetic_accuracy": 0.9990950226244344,
|
712 |
+
"eval_gptsynthetic_loss": 0.0032840222120285034,
|
713 |
+
"eval_gptsynthetic_runtime": 101.8215,
|
714 |
+
"eval_gptsynthetic_samples_per_second": 32.557,
|
715 |
+
"eval_gptsynthetic_steps_per_second": 6.511,
|
716 |
+
"step": 1000
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.53,
|
720 |
+
"learning_rate": 9.438815009955701e-06,
|
721 |
+
"loss": 0.386,
|
722 |
+
"step": 1010
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 0.53,
|
726 |
+
"learning_rate": 9.41699088313398e-06,
|
727 |
+
"loss": 0.401,
|
728 |
+
"step": 1020
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 0.54,
|
732 |
+
"learning_rate": 9.394776679352675e-06,
|
733 |
+
"loss": 0.4199,
|
734 |
+
"step": 1030
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 0.54,
|
738 |
+
"learning_rate": 9.372174360410963e-06,
|
739 |
+
"loss": 0.4093,
|
740 |
+
"step": 1040
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 0.55,
|
744 |
+
"learning_rate": 9.349185922383564e-06,
|
745 |
+
"loss": 0.4313,
|
746 |
+
"step": 1050
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 0.55,
|
750 |
+
"learning_rate": 9.325813395444472e-06,
|
751 |
+
"loss": 0.4164,
|
752 |
+
"step": 1060
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 0.56,
|
756 |
+
"learning_rate": 9.302058843687655e-06,
|
757 |
+
"loss": 0.3871,
|
758 |
+
"step": 1070
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.56,
|
762 |
+
"learning_rate": 9.277924364944774e-06,
|
763 |
+
"loss": 0.4115,
|
764 |
+
"step": 1080
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 0.57,
|
768 |
+
"learning_rate": 9.253412090599923e-06,
|
769 |
+
"loss": 0.4094,
|
770 |
+
"step": 1090
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 0.57,
|
774 |
+
"learning_rate": 9.228524185401384e-06,
|
775 |
+
"loss": 0.3957,
|
776 |
+
"step": 1100
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 0.58,
|
780 |
+
"learning_rate": 9.203262847270469e-06,
|
781 |
+
"loss": 0.3952,
|
782 |
+
"step": 1110
|
783 |
+
},
|
784 |
+
{
|
785 |
+
"epoch": 0.58,
|
786 |
+
"learning_rate": 9.177630307107408e-06,
|
787 |
+
"loss": 0.4011,
|
788 |
+
"step": 1120
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 0.59,
|
792 |
+
"learning_rate": 9.15162882859433e-06,
|
793 |
+
"loss": 0.4042,
|
794 |
+
"step": 1130
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 0.59,
|
798 |
+
"learning_rate": 9.125260707995356e-06,
|
799 |
+
"loss": 0.404,
|
800 |
+
"step": 1140
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.6,
|
804 |
+
"learning_rate": 9.098528273953807e-06,
|
805 |
+
"loss": 0.3904,
|
806 |
+
"step": 1150
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 0.6,
|
810 |
+
"learning_rate": 9.071433887286551e-06,
|
811 |
+
"loss": 0.4011,
|
812 |
+
"step": 1160
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 0.61,
|
816 |
+
"learning_rate": 9.04397994077552e-06,
|
817 |
+
"loss": 0.3806,
|
818 |
+
"step": 1170
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 0.61,
|
822 |
+
"learning_rate": 9.01616885895639e-06,
|
823 |
+
"loss": 0.3878,
|
824 |
+
"step": 1180
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 0.62,
|
828 |
+
"learning_rate": 8.988003097904472e-06,
|
829 |
+
"loss": 0.384,
|
830 |
+
"step": 1190
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 0.62,
|
834 |
+
"learning_rate": 8.959485145017794e-06,
|
835 |
+
"loss": 0.4023,
|
836 |
+
"step": 1200
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.63,
|
840 |
+
"learning_rate": 8.930617518797448e-06,
|
841 |
+
"loss": 0.3609,
|
842 |
+
"step": 1210
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.63,
|
846 |
+
"learning_rate": 8.901402768625169e-06,
|
847 |
+
"loss": 0.3879,
|
848 |
+
"step": 1220
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 0.64,
|
852 |
+
"learning_rate": 8.871843474538182e-06,
|
853 |
+
"loss": 0.4003,
|
854 |
+
"step": 1230
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 0.65,
|
858 |
+
"learning_rate": 8.841942247001368e-06,
|
859 |
+
"loss": 0.3974,
|
860 |
+
"step": 1240
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 0.65,
|
864 |
+
"learning_rate": 8.811701726676711e-06,
|
865 |
+
"loss": 0.4067,
|
866 |
+
"step": 1250
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 0.65,
|
870 |
+
"eval_webgpt_accuracy": 0.5967824310520939,
|
871 |
+
"eval_webgpt_loss": 0.6645339727401733,
|
872 |
+
"eval_webgpt_runtime": 167.52,
|
873 |
+
"eval_webgpt_samples_per_second": 23.376,
|
874 |
+
"eval_webgpt_steps_per_second": 4.68,
|
875 |
+
"step": 1250
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 0.65,
|
879 |
+
"eval_hfsummary_accuracy": 0.7167125109572893,
|
880 |
+
"eval_hfsummary_loss": 0.559929370880127,
|
881 |
+
"eval_hfsummary_runtime": 3290.2163,
|
882 |
+
"eval_hfsummary_samples_per_second": 10.055,
|
883 |
+
"eval_hfsummary_steps_per_second": 2.011,
|
884 |
+
"step": 1250
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.65,
|
888 |
+
"eval_gptsynthetic_accuracy": 0.9993966817496229,
|
889 |
+
"eval_gptsynthetic_loss": 0.002522215712815523,
|
890 |
+
"eval_gptsynthetic_runtime": 116.8538,
|
891 |
+
"eval_gptsynthetic_samples_per_second": 28.369,
|
892 |
+
"eval_gptsynthetic_steps_per_second": 5.674,
|
893 |
+
"step": 1250
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 0.66,
|
897 |
+
"learning_rate": 8.78112458419011e-06,
|
898 |
+
"loss": 0.3581,
|
899 |
+
"step": 1260
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 0.66,
|
903 |
+
"learning_rate": 8.750213519895511e-06,
|
904 |
+
"loss": 0.3628,
|
905 |
+
"step": 1270
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.67,
|
909 |
+
"learning_rate": 8.718971263636441e-06,
|
910 |
+
"loss": 0.3799,
|
911 |
+
"step": 1280
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 0.67,
|
915 |
+
"learning_rate": 8.687400574504929e-06,
|
916 |
+
"loss": 0.4083,
|
917 |
+
"step": 1290
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 0.68,
|
921 |
+
"learning_rate": 8.655504240597841e-06,
|
922 |
+
"loss": 0.3771,
|
923 |
+
"step": 1300
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 0.68,
|
927 |
+
"learning_rate": 8.623285078770649e-06,
|
928 |
+
"loss": 0.4016,
|
929 |
+
"step": 1310
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 0.69,
|
933 |
+
"learning_rate": 8.590745934388675e-06,
|
934 |
+
"loss": 0.3878,
|
935 |
+
"step": 1320
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 0.69,
|
939 |
+
"learning_rate": 8.557889681075808e-06,
|
940 |
+
"loss": 0.3568,
|
941 |
+
"step": 1330
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 0.7,
|
945 |
+
"learning_rate": 8.524719220460718e-06,
|
946 |
+
"loss": 0.3914,
|
947 |
+
"step": 1340
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.7,
|
951 |
+
"learning_rate": 8.491237481920616e-06,
|
952 |
+
"loss": 0.4001,
|
953 |
+
"step": 1350
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 0.71,
|
957 |
+
"learning_rate": 8.457447422322545e-06,
|
958 |
+
"loss": 0.3683,
|
959 |
+
"step": 1360
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 0.71,
|
963 |
+
"learning_rate": 8.42335202576225e-06,
|
964 |
+
"loss": 0.3741,
|
965 |
+
"step": 1370
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 0.72,
|
969 |
+
"learning_rate": 8.388954303300651e-06,
|
970 |
+
"loss": 0.3636,
|
971 |
+
"step": 1380
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 0.72,
|
975 |
+
"learning_rate": 8.354257292697927e-06,
|
976 |
+
"loss": 0.3815,
|
977 |
+
"step": 1390
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 0.73,
|
981 |
+
"learning_rate": 8.319264058145227e-06,
|
982 |
+
"loss": 0.3596,
|
983 |
+
"step": 1400
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 0.73,
|
987 |
+
"learning_rate": 8.283977689994085e-06,
|
988 |
+
"loss": 0.3826,
|
989 |
+
"step": 1410
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.74,
|
993 |
+
"learning_rate": 8.248401304483492e-06,
|
994 |
+
"loss": 0.41,
|
995 |
+
"step": 1420
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 0.74,
|
999 |
+
"learning_rate": 8.21253804346469e-06,
|
1000 |
+
"loss": 0.3648,
|
1001 |
+
"step": 1430
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 0.75,
|
1005 |
+
"learning_rate": 8.176391074123705e-06,
|
1006 |
+
"loss": 0.3439,
|
1007 |
+
"step": 1440
|
1008 |
+
},
|
1009 |
+
{
|
1010 |
+
"epoch": 0.75,
|
1011 |
+
"learning_rate": 8.139963588701656e-06,
|
1012 |
+
"loss": 0.3635,
|
1013 |
+
"step": 1450
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 0.76,
|
1017 |
+
"learning_rate": 8.103258804212824e-06,
|
1018 |
+
"loss": 0.3633,
|
1019 |
+
"step": 1460
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 0.77,
|
1023 |
+
"learning_rate": 8.066279962160554e-06,
|
1024 |
+
"loss": 0.3989,
|
1025 |
+
"step": 1470
|
1026 |
+
},
|
1027 |
+
{
|
1028 |
+
"epoch": 0.77,
|
1029 |
+
"learning_rate": 8.029030328250997e-06,
|
1030 |
+
"loss": 0.3504,
|
1031 |
+
"step": 1480
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.78,
|
1035 |
+
"learning_rate": 7.991513192104694e-06,
|
1036 |
+
"loss": 0.3665,
|
1037 |
+
"step": 1490
|
1038 |
+
},
|
1039 |
+
{
|
1040 |
+
"epoch": 0.78,
|
1041 |
+
"learning_rate": 7.953731866966063e-06,
|
1042 |
+
"loss": 0.3874,
|
1043 |
+
"step": 1500
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 0.78,
|
1047 |
+
"eval_webgpt_accuracy": 0.6113381001021451,
|
1048 |
+
"eval_webgpt_loss": 0.6596858501434326,
|
1049 |
+
"eval_webgpt_runtime": 179.0732,
|
1050 |
+
"eval_webgpt_samples_per_second": 21.868,
|
1051 |
+
"eval_webgpt_steps_per_second": 4.378,
|
1052 |
+
"step": 1500
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.78,
|
1056 |
+
"eval_hfsummary_accuracy": 0.7207327025964997,
|
1057 |
+
"eval_hfsummary_loss": 0.5737717747688293,
|
1058 |
+
"eval_hfsummary_runtime": 3606.3862,
|
1059 |
+
"eval_hfsummary_samples_per_second": 9.173,
|
1060 |
+
"eval_hfsummary_steps_per_second": 1.835,
|
1061 |
+
"step": 1500
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 0.78,
|
1065 |
+
"eval_gptsynthetic_accuracy": 0.9993966817496229,
|
1066 |
+
"eval_gptsynthetic_loss": 0.002385695930570364,
|
1067 |
+
"eval_gptsynthetic_runtime": 109.239,
|
1068 |
+
"eval_gptsynthetic_samples_per_second": 30.346,
|
1069 |
+
"eval_gptsynthetic_steps_per_second": 6.069,
|
1070 |
+
"step": 1500
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 0.79,
|
1074 |
+
"learning_rate": 7.915689689410806e-06,
|
1075 |
+
"loss": 0.3532,
|
1076 |
+
"step": 1510
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 0.79,
|
1080 |
+
"learning_rate": 7.877390019051232e-06,
|
1081 |
+
"loss": 0.403,
|
1082 |
+
"step": 1520
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 0.8,
|
1086 |
+
"learning_rate": 7.838836238239574e-06,
|
1087 |
+
"loss": 0.3735,
|
1088 |
+
"step": 1530
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 0.8,
|
1092 |
+
"learning_rate": 7.800031751769274e-06,
|
1093 |
+
"loss": 0.3773,
|
1094 |
+
"step": 1540
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.81,
|
1098 |
+
"learning_rate": 7.760979986574304e-06,
|
1099 |
+
"loss": 0.3644,
|
1100 |
+
"step": 1550
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 0.81,
|
1104 |
+
"learning_rate": 7.721684391426516e-06,
|
1105 |
+
"loss": 0.3692,
|
1106 |
+
"step": 1560
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 0.82,
|
1110 |
+
"learning_rate": 7.682148436631075e-06,
|
1111 |
+
"loss": 0.3804,
|
1112 |
+
"step": 1570
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 0.82,
|
1116 |
+
"learning_rate": 7.642375613719988e-06,
|
1117 |
+
"loss": 0.397,
|
1118 |
+
"step": 1580
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 0.83,
|
1122 |
+
"learning_rate": 7.602369435143755e-06,
|
1123 |
+
"loss": 0.3842,
|
1124 |
+
"step": 1590
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 0.83,
|
1128 |
+
"learning_rate": 7.562133433961176e-06,
|
1129 |
+
"loss": 0.3571,
|
1130 |
+
"step": 1600
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 0.84,
|
1134 |
+
"learning_rate": 7.5216711635273305e-06,
|
1135 |
+
"loss": 0.3784,
|
1136 |
+
"step": 1610
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.84,
|
1140 |
+
"learning_rate": 7.480986197179785e-06,
|
1141 |
+
"loss": 0.3947,
|
1142 |
+
"step": 1620
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 0.85,
|
1146 |
+
"learning_rate": 7.440082127923002e-06,
|
1147 |
+
"loss": 0.3583,
|
1148 |
+
"step": 1630
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 0.85,
|
1152 |
+
"learning_rate": 7.39896256811105e-06,
|
1153 |
+
"loss": 0.3671,
|
1154 |
+
"step": 1640
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 0.86,
|
1158 |
+
"learning_rate": 7.357631149128568e-06,
|
1159 |
+
"loss": 0.3691,
|
1160 |
+
"step": 1650
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 0.86,
|
1164 |
+
"learning_rate": 7.31609152107009e-06,
|
1165 |
+
"loss": 0.4069,
|
1166 |
+
"step": 1660
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 0.87,
|
1170 |
+
"learning_rate": 7.2743473524176734e-06,
|
1171 |
+
"loss": 0.3515,
|
1172 |
+
"step": 1670
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 0.87,
|
1176 |
+
"learning_rate": 7.232402329716938e-06,
|
1177 |
+
"loss": 0.3783,
|
1178 |
+
"step": 1680
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.88,
|
1182 |
+
"learning_rate": 7.19026015725149e-06,
|
1183 |
+
"loss": 0.362,
|
1184 |
+
"step": 1690
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 0.88,
|
1188 |
+
"learning_rate": 7.147924556715793e-06,
|
1189 |
+
"loss": 0.3807,
|
1190 |
+
"step": 1700
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 0.89,
|
1194 |
+
"learning_rate": 7.1053992668864855e-06,
|
1195 |
+
"loss": 0.3632,
|
1196 |
+
"step": 1710
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 0.9,
|
1200 |
+
"learning_rate": 7.0626880432922095e-06,
|
1201 |
+
"loss": 0.3619,
|
1202 |
+
"step": 1720
|
1203 |
+
},
|
1204 |
+
{
|
1205 |
+
"epoch": 0.9,
|
1206 |
+
"learning_rate": 7.019794657881942e-06,
|
1207 |
+
"loss": 0.3883,
|
1208 |
+
"step": 1730
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 0.91,
|
1212 |
+
"learning_rate": 6.976722898691881e-06,
|
1213 |
+
"loss": 0.351,
|
1214 |
+
"step": 1740
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 0.91,
|
1218 |
+
"learning_rate": 6.933476569510924e-06,
|
1219 |
+
"loss": 0.3613,
|
1220 |
+
"step": 1750
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.91,
|
1224 |
+
"eval_webgpt_accuracy": 0.609805924412666,
|
1225 |
+
"eval_webgpt_loss": 0.6571772694587708,
|
1226 |
+
"eval_webgpt_runtime": 167.3559,
|
1227 |
+
"eval_webgpt_samples_per_second": 23.399,
|
1228 |
+
"eval_webgpt_steps_per_second": 4.685,
|
1229 |
+
"step": 1750
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 0.91,
|
1233 |
+
"eval_hfsummary_accuracy": 0.7195538494090621,
|
1234 |
+
"eval_hfsummary_loss": 0.5861465334892273,
|
1235 |
+
"eval_hfsummary_runtime": 3235.9038,
|
1236 |
+
"eval_hfsummary_samples_per_second": 10.224,
|
1237 |
+
"eval_hfsummary_steps_per_second": 2.045,
|
1238 |
+
"step": 1750
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 0.91,
|
1242 |
+
"eval_gptsynthetic_accuracy": 0.9996983408748115,
|
1243 |
+
"eval_gptsynthetic_loss": 0.0027254121378064156,
|
1244 |
+
"eval_gptsynthetic_runtime": 101.6222,
|
1245 |
+
"eval_gptsynthetic_samples_per_second": 32.621,
|
1246 |
+
"eval_gptsynthetic_steps_per_second": 6.524,
|
1247 |
+
"step": 1750
|
1248 |
+
},
|
1249 |
+
{
|
1250 |
+
"epoch": 0.92,
|
1251 |
+
"learning_rate": 6.890059489544735e-06,
|
1252 |
+
"loss": 0.3651,
|
1253 |
+
"step": 1760
|
1254 |
+
},
|
1255 |
+
{
|
1256 |
+
"epoch": 0.92,
|
1257 |
+
"learning_rate": 6.846475493078462e-06,
|
1258 |
+
"loss": 0.3617,
|
1259 |
+
"step": 1770
|
1260 |
+
},
|
1261 |
+
{
|
1262 |
+
"epoch": 0.93,
|
1263 |
+
"learning_rate": 6.802728429138122e-06,
|
1264 |
+
"loss": 0.3174,
|
1265 |
+
"step": 1780
|
1266 |
+
},
|
1267 |
+
{
|
1268 |
+
"epoch": 0.93,
|
1269 |
+
"learning_rate": 6.758822161150687e-06,
|
1270 |
+
"loss": 0.3399,
|
1271 |
+
"step": 1790
|
1272 |
+
},
|
1273 |
+
{
|
1274 |
+
"epoch": 0.94,
|
1275 |
+
"learning_rate": 6.714760566602876e-06,
|
1276 |
+
"loss": 0.3664,
|
1277 |
+
"step": 1800
|
1278 |
+
},
|
1279 |
+
{
|
1280 |
+
"epoch": 0.94,
|
1281 |
+
"learning_rate": 6.67054753669875e-06,
|
1282 |
+
"loss": 0.3432,
|
1283 |
+
"step": 1810
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.95,
|
1287 |
+
"learning_rate": 6.626186976016049e-06,
|
1288 |
+
"loss": 0.3673,
|
1289 |
+
"step": 1820
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 0.95,
|
1293 |
+
"learning_rate": 6.581682802161371e-06,
|
1294 |
+
"loss": 0.3562,
|
1295 |
+
"step": 1830
|
1296 |
+
},
|
1297 |
+
{
|
1298 |
+
"epoch": 0.96,
|
1299 |
+
"learning_rate": 6.537038945424201e-06,
|
1300 |
+
"loss": 0.338,
|
1301 |
+
"step": 1840
|
1302 |
+
},
|
1303 |
+
{
|
1304 |
+
"epoch": 0.96,
|
1305 |
+
"learning_rate": 6.492259348429818e-06,
|
1306 |
+
"loss": 0.3387,
|
1307 |
+
"step": 1850
|
1308 |
+
},
|
1309 |
+
{
|
1310 |
+
"epoch": 0.97,
|
1311 |
+
"learning_rate": 6.4473479657910995e-06,
|
1312 |
+
"loss": 0.3504,
|
1313 |
+
"step": 1860
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 0.97,
|
1317 |
+
"learning_rate": 6.402308763759296e-06,
|
1318 |
+
"loss": 0.3276,
|
1319 |
+
"step": 1870
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 0.98,
|
1323 |
+
"learning_rate": 6.357145719873742e-06,
|
1324 |
+
"loss": 0.3243,
|
1325 |
+
"step": 1880
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.98,
|
1329 |
+
"learning_rate": 6.3118628226105995e-06,
|
1330 |
+
"loss": 0.3548,
|
1331 |
+
"step": 1890
|
1332 |
+
},
|
1333 |
+
{
|
1334 |
+
"epoch": 0.99,
|
1335 |
+
"learning_rate": 6.266464071030621e-06,
|
1336 |
+
"loss": 0.3099,
|
1337 |
+
"step": 1900
|
1338 |
+
},
|
1339 |
+
{
|
1340 |
+
"epoch": 0.99,
|
1341 |
+
"learning_rate": 6.220953474425984e-06,
|
1342 |
+
"loss": 0.3387,
|
1343 |
+
"step": 1910
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 1.0,
|
1347 |
+
"learning_rate": 6.175335051966211e-06,
|
1348 |
+
"loss": 0.3403,
|
1349 |
+
"step": 1920
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 1.0,
|
1353 |
+
"learning_rate": 6.129612832343233e-06,
|
1354 |
+
"loss": 0.3385,
|
1355 |
+
"step": 1930
|
1356 |
+
},
|
1357 |
+
{
|
1358 |
+
"epoch": 1.01,
|
1359 |
+
"learning_rate": 6.0837908534156076e-06,
|
1360 |
+
"loss": 0.3305,
|
1361 |
+
"step": 1940
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 1.02,
|
1365 |
+
"learning_rate": 6.037873161851911e-06,
|
1366 |
+
"loss": 0.3421,
|
1367 |
+
"step": 1950
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 1.02,
|
1371 |
+
"learning_rate": 5.9918638127733764e-06,
|
1372 |
+
"loss": 0.3381,
|
1373 |
+
"step": 1960
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 1.03,
|
1377 |
+
"learning_rate": 5.945766869395772e-06,
|
1378 |
+
"loss": 0.3062,
|
1379 |
+
"step": 1970
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 1.03,
|
1383 |
+
"learning_rate": 5.899586402670566e-06,
|
1384 |
+
"loss": 0.3459,
|
1385 |
+
"step": 1980
|
1386 |
+
},
|
1387 |
+
{
|
1388 |
+
"epoch": 1.04,
|
1389 |
+
"learning_rate": 5.853326490925407e-06,
|
1390 |
+
"loss": 0.3175,
|
1391 |
+
"step": 1990
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 1.04,
|
1395 |
+
"learning_rate": 5.806991219503963e-06,
|
1396 |
+
"loss": 0.3117,
|
1397 |
+
"step": 2000
|
1398 |
+
},
|
1399 |
+
{
|
1400 |
+
"epoch": 1.04,
|
1401 |
+
"eval_webgpt_accuracy": 0.6046986721144024,
|
1402 |
+
"eval_webgpt_loss": 0.6649239659309387,
|
1403 |
+
"eval_webgpt_runtime": 166.8662,
|
1404 |
+
"eval_webgpt_samples_per_second": 23.468,
|
1405 |
+
"eval_webgpt_steps_per_second": 4.698,
|
1406 |
+
"step": 2000
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 1.04,
|
1410 |
+
"eval_hfsummary_accuracy": 0.7152616147265968,
|
1411 |
+
"eval_hfsummary_loss": 0.6051351428031921,
|
1412 |
+
"eval_hfsummary_runtime": 3227.7808,
|
1413 |
+
"eval_hfsummary_samples_per_second": 10.249,
|
1414 |
+
"eval_hfsummary_steps_per_second": 2.05,
|
1415 |
+
"step": 2000
|
1416 |
+
},
|
1417 |
+
{
|
1418 |
+
"epoch": 1.04,
|
1419 |
+
"eval_gptsynthetic_accuracy": 1.0,
|
1420 |
+
"eval_gptsynthetic_loss": 0.002014968078583479,
|
1421 |
+
"eval_gptsynthetic_runtime": 101.558,
|
1422 |
+
"eval_gptsynthetic_samples_per_second": 32.641,
|
1423 |
+
"eval_gptsynthetic_steps_per_second": 6.528,
|
1424 |
+
"step": 2000
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 1.05,
|
1428 |
+
"learning_rate": 5.765228424606831e-06,
|
1429 |
+
"loss": 0.3206,
|
1430 |
+
"step": 2010
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 1.05,
|
1434 |
+
"learning_rate": 5.718761248466732e-06,
|
1435 |
+
"loss": 0.3073,
|
1436 |
+
"step": 2020
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.06,
|
1440 |
+
"learning_rate": 5.672230596487563e-06,
|
1441 |
+
"loss": 0.3113,
|
1442 |
+
"step": 2030
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 1.06,
|
1446 |
+
"learning_rate": 5.625640577922828e-06,
|
1447 |
+
"loss": 0.3213,
|
1448 |
+
"step": 2040
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 1.07,
|
1452 |
+
"learning_rate": 5.578995307268855e-06,
|
1453 |
+
"loss": 0.3353,
|
1454 |
+
"step": 2050
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 1.07,
|
1458 |
+
"learning_rate": 5.532298903901448e-06,
|
1459 |
+
"loss": 0.3293,
|
1460 |
+
"step": 2060
|
1461 |
+
},
|
1462 |
+
{
|
1463 |
+
"epoch": 1.08,
|
1464 |
+
"learning_rate": 5.485555491712082e-06,
|
1465 |
+
"loss": 0.3158,
|
1466 |
+
"step": 2070
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 1.08,
|
1470 |
+
"learning_rate": 5.438769198743713e-06,
|
1471 |
+
"loss": 0.2968,
|
1472 |
+
"step": 2080
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 1.09,
|
1476 |
+
"learning_rate": 5.391944156826221e-06,
|
1477 |
+
"loss": 0.2877,
|
1478 |
+
"step": 2090
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 1.09,
|
1482 |
+
"learning_rate": 5.345084501211517e-06,
|
1483 |
+
"loss": 0.3361,
|
1484 |
+
"step": 2100
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 1.1,
|
1488 |
+
"learning_rate": 5.298194370208341e-06,
|
1489 |
+
"loss": 0.3051,
|
1490 |
+
"step": 2110
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 1.1,
|
1494 |
+
"learning_rate": 5.251277904816809e-06,
|
1495 |
+
"loss": 0.3398,
|
1496 |
+
"step": 2120
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 1.11,
|
1500 |
+
"learning_rate": 5.2043392483626906e-06,
|
1501 |
+
"loss": 0.3155,
|
1502 |
+
"step": 2130
|
1503 |
+
},
|
1504 |
+
{
|
1505 |
+
"epoch": 1.11,
|
1506 |
+
"learning_rate": 5.157382546131519e-06,
|
1507 |
+
"loss": 0.315,
|
1508 |
+
"step": 2140
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 1.12,
|
1512 |
+
"learning_rate": 5.110411945002497e-06,
|
1513 |
+
"loss": 0.304,
|
1514 |
+
"step": 2150
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 1.12,
|
1518 |
+
"learning_rate": 5.063431593082283e-06,
|
1519 |
+
"loss": 0.3458,
|
1520 |
+
"step": 2160
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 1.13,
|
1524 |
+
"learning_rate": 5.016445639338647e-06,
|
1525 |
+
"loss": 0.3192,
|
1526 |
+
"step": 2170
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 1.13,
|
1530 |
+
"learning_rate": 4.969458233234083e-06,
|
1531 |
+
"loss": 0.3188,
|
1532 |
+
"step": 2180
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 1.14,
|
1536 |
+
"learning_rate": 4.922473524359335e-06,
|
1537 |
+
"loss": 0.308,
|
1538 |
+
"step": 2190
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 1.15,
|
1542 |
+
"learning_rate": 4.875495662066958e-06,
|
1543 |
+
"loss": 0.3052,
|
1544 |
+
"step": 2200
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 1.15,
|
1548 |
+
"learning_rate": 4.82852879510486e-06,
|
1549 |
+
"loss": 0.2962,
|
1550 |
+
"step": 2210
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 1.16,
|
1554 |
+
"learning_rate": 4.781577071249918e-06,
|
1555 |
+
"loss": 0.3067,
|
1556 |
+
"step": 2220
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 1.16,
|
1560 |
+
"learning_rate": 4.734644636941687e-06,
|
1561 |
+
"loss": 0.2886,
|
1562 |
+
"step": 2230
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 1.17,
|
1566 |
+
"learning_rate": 4.6877356369161975e-06,
|
1567 |
+
"loss": 0.2832,
|
1568 |
+
"step": 2240
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 1.17,
|
1572 |
+
"learning_rate": 4.64085421383994e-06,
|
1573 |
+
"loss": 0.3177,
|
1574 |
+
"step": 2250
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 1.17,
|
1578 |
+
"eval_webgpt_accuracy": 0.6052093973442288,
|
1579 |
+
"eval_webgpt_loss": 0.67301344871521,
|
1580 |
+
"eval_webgpt_runtime": 167.0393,
|
1581 |
+
"eval_webgpt_samples_per_second": 23.444,
|
1582 |
+
"eval_webgpt_steps_per_second": 4.694,
|
1583 |
+
"step": 2250
|
1584 |
+
},
|
1585 |
+
{
|
1586 |
+
"epoch": 1.17,
|
1587 |
+
"eval_hfsummary_accuracy": 0.7188586282985219,
|
1588 |
+
"eval_hfsummary_loss": 0.665126621723175,
|
1589 |
+
"eval_hfsummary_runtime": 3228.0014,
|
1590 |
+
"eval_hfsummary_samples_per_second": 10.249,
|
1591 |
+
"eval_hfsummary_steps_per_second": 2.05,
|
1592 |
+
"step": 2250
|
1593 |
+
},
|
1594 |
+
{
|
1595 |
+
"epoch": 1.17,
|
1596 |
+
"eval_gptsynthetic_accuracy": 1.0,
|
1597 |
+
"eval_gptsynthetic_loss": 0.0015043159946799278,
|
1598 |
+
"eval_gptsynthetic_runtime": 101.5554,
|
1599 |
+
"eval_gptsynthetic_samples_per_second": 32.642,
|
1600 |
+
"eval_gptsynthetic_steps_per_second": 6.528,
|
1601 |
+
"step": 2250
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 1.18,
|
1605 |
+
"learning_rate": 4.594004507944001e-06,
|
1606 |
+
"loss": 0.2988,
|
1607 |
+
"step": 2260
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 1.18,
|
1611 |
+
"learning_rate": 4.547190656658441e-06,
|
1612 |
+
"loss": 0.3037,
|
1613 |
+
"step": 2270
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 1.19,
|
1617 |
+
"learning_rate": 4.500416794246885e-06,
|
1618 |
+
"loss": 0.3024,
|
1619 |
+
"step": 2280
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 1.19,
|
1623 |
+
"learning_rate": 4.453687051441443e-06,
|
1624 |
+
"loss": 0.3008,
|
1625 |
+
"step": 2290
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 1.2,
|
1629 |
+
"learning_rate": 4.407005555077885e-06,
|
1630 |
+
"loss": 0.3059,
|
1631 |
+
"step": 2300
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 1.2,
|
1635 |
+
"learning_rate": 4.36037642773121e-06,
|
1636 |
+
"loss": 0.3191,
|
1637 |
+
"step": 2310
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 1.21,
|
1641 |
+
"learning_rate": 4.313803787351555e-06,
|
1642 |
+
"loss": 0.3228,
|
1643 |
+
"step": 2320
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 1.21,
|
1647 |
+
"learning_rate": 4.267291746900538e-06,
|
1648 |
+
"loss": 0.3054,
|
1649 |
+
"step": 2330
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 1.22,
|
1653 |
+
"learning_rate": 4.220844413988022e-06,
|
1654 |
+
"loss": 0.3161,
|
1655 |
+
"step": 2340
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 1.22,
|
1659 |
+
"learning_rate": 4.174465890509365e-06,
|
1660 |
+
"loss": 0.296,
|
1661 |
+
"step": 2350
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 1.23,
|
1665 |
+
"learning_rate": 4.128160272283172e-06,
|
1666 |
+
"loss": 0.2718,
|
1667 |
+
"step": 2360
|
1668 |
+
},
|
1669 |
+
{
|
1670 |
+
"epoch": 1.23,
|
1671 |
+
"learning_rate": 4.081931648689577e-06,
|
1672 |
+
"loss": 0.3204,
|
1673 |
+
"step": 2370
|
1674 |
+
},
|
1675 |
+
{
|
1676 |
+
"epoch": 1.24,
|
1677 |
+
"learning_rate": 4.035784102309097e-06,
|
1678 |
+
"loss": 0.3418,
|
1679 |
+
"step": 2380
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 1.24,
|
1683 |
+
"learning_rate": 3.989721708562098e-06,
|
1684 |
+
"loss": 0.3246,
|
1685 |
+
"step": 2390
|
1686 |
+
},
|
1687 |
+
{
|
1688 |
+
"epoch": 1.25,
|
1689 |
+
"learning_rate": 3.943748535348872e-06,
|
1690 |
+
"loss": 0.2989,
|
1691 |
+
"step": 2400
|
1692 |
+
},
|
1693 |
+
{
|
1694 |
+
"epoch": 1.25,
|
1695 |
+
"learning_rate": 3.8978686426903885e-06,
|
1696 |
+
"loss": 0.3318,
|
1697 |
+
"step": 2410
|
1698 |
+
},
|
1699 |
+
{
|
1700 |
+
"epoch": 1.26,
|
1701 |
+
"learning_rate": 3.852086082369758e-06,
|
1702 |
+
"loss": 0.2856,
|
1703 |
+
"step": 2420
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 1.26,
|
1707 |
+
"learning_rate": 3.806404897574388e-06,
|
1708 |
+
"loss": 0.2934,
|
1709 |
+
"step": 2430
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 1.27,
|
1713 |
+
"learning_rate": 3.760829122538931e-06,
|
1714 |
+
"loss": 0.3017,
|
1715 |
+
"step": 2440
|
1716 |
+
},
|
1717 |
+
{
|
1718 |
+
"epoch": 1.28,
|
1719 |
+
"learning_rate": 3.7153627821890026e-06,
|
1720 |
+
"loss": 0.318,
|
1721 |
+
"step": 2450
|
1722 |
+
},
|
1723 |
+
{
|
1724 |
+
"epoch": 1.28,
|
1725 |
+
"learning_rate": 3.6700098917857387e-06,
|
1726 |
+
"loss": 0.2964,
|
1727 |
+
"step": 2460
|
1728 |
+
},
|
1729 |
+
{
|
1730 |
+
"epoch": 1.29,
|
1731 |
+
"learning_rate": 3.624774456571176e-06,
|
1732 |
+
"loss": 0.2909,
|
1733 |
+
"step": 2470
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 1.29,
|
1737 |
+
"learning_rate": 3.579660471414565e-06,
|
1738 |
+
"loss": 0.3043,
|
1739 |
+
"step": 2480
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 1.3,
|
1743 |
+
"learning_rate": 3.5346719204595497e-06,
|
1744 |
+
"loss": 0.305,
|
1745 |
+
"step": 2490
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 1.3,
|
1749 |
+
"learning_rate": 3.48981277677233e-06,
|
1750 |
+
"loss": 0.2756,
|
1751 |
+
"step": 2500
|
1752 |
+
},
|
1753 |
+
{
|
1754 |
+
"epoch": 1.3,
|
1755 |
+
"eval_webgpt_accuracy": 0.6131256384065373,
|
1756 |
+
"eval_webgpt_loss": 0.6625245809555054,
|
1757 |
+
"eval_webgpt_runtime": 166.7899,
|
1758 |
+
"eval_webgpt_samples_per_second": 23.479,
|
1759 |
+
"eval_webgpt_steps_per_second": 4.701,
|
1760 |
+
"step": 2500
|
1761 |
+
},
|
1762 |
+
{
|
1763 |
+
"epoch": 1.3,
|
1764 |
+
"eval_hfsummary_accuracy": 0.7223347338512227,
|
1765 |
+
"eval_hfsummary_loss": 0.6251139044761658,
|
1766 |
+
"eval_hfsummary_runtime": 3221.6572,
|
1767 |
+
"eval_hfsummary_samples_per_second": 10.269,
|
1768 |
+
"eval_hfsummary_steps_per_second": 2.054,
|
1769 |
+
"step": 2500
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 1.3,
|
1773 |
+
"eval_gptsynthetic_accuracy": 0.9993966817496229,
|
1774 |
+
"eval_gptsynthetic_loss": 0.00219121971167624,
|
1775 |
+
"eval_gptsynthetic_runtime": 101.3035,
|
1776 |
+
"eval_gptsynthetic_samples_per_second": 32.723,
|
1777 |
+
"eval_gptsynthetic_steps_per_second": 6.545,
|
1778 |
+
"step": 2500
|
1779 |
+
}
|
1780 |
+
],
|
1781 |
+
"max_steps": 3842,
|
1782 |
+
"num_train_epochs": 2,
|
1783 |
+
"total_flos": 4.2100836618337517e+17,
|
1784 |
+
"trial_name": null,
|
1785 |
+
"trial_params": null
|
1786 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b3b682c3419f7b6258e04c56902b36cd4b54c4c8ff5d1f32213cbe7a011b6375
|
3 |
+
size 3439
|