sebastianhestsveen
commited on
Commit
•
9c241e6
1
Parent(s):
684293a
Upload 10 files
Browse files- config.json +32 -0
- generation_config.json +7 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +5 -0
- spiece.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +12 -0
- trainer_state.json +4018 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/mnt/batch/tasks/shared/LS_root/mounts/clusters/sebhes1/code/Users/erlhje/mt5-large-pruned",
|
3 |
+
"architectures": [
|
4 |
+
"MT5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"d_ff": 2816,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 1024,
|
9 |
+
"decoder_start_token_id": 0,
|
10 |
+
"dense_act_fn": "gelu_new",
|
11 |
+
"dropout_rate": 0.1,
|
12 |
+
"eos_token_id": 1,
|
13 |
+
"feed_forward_proj": "gated-gelu",
|
14 |
+
"initializer_factor": 1.0,
|
15 |
+
"is_encoder_decoder": true,
|
16 |
+
"is_gated_act": true,
|
17 |
+
"layer_norm_epsilon": 1e-06,
|
18 |
+
"model_type": "mt5",
|
19 |
+
"num_decoder_layers": 24,
|
20 |
+
"num_heads": 16,
|
21 |
+
"num_layers": 24,
|
22 |
+
"output_past": true,
|
23 |
+
"pad_token_id": 0,
|
24 |
+
"relative_attention_max_distance": 128,
|
25 |
+
"relative_attention_num_buckets": 32,
|
26 |
+
"tie_word_embeddings": false,
|
27 |
+
"tokenizer_class": "T5Tokenizer",
|
28 |
+
"torch_dtype": "float32",
|
29 |
+
"transformers_version": "4.31.0",
|
30 |
+
"use_cache": true,
|
31 |
+
"vocab_size": 30000
|
32 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"decoder_start_token_id": 0,
|
4 |
+
"eos_token_id": 1,
|
5 |
+
"pad_token_id": 0,
|
6 |
+
"transformers_version": "4.31.0"
|
7 |
+
}
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8fb3c2a0207824c015d8574cb83fb4ba24e6eed8381855bfb3876d3d041f7d1f
|
3 |
+
size 13611
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6086e80de9a299e8ebc9c0d037531136a1d7f8bb7c5020275636e6f38cde4e15
|
3 |
+
size 623
|
special_tokens_map.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eos_token": "</s>",
|
3 |
+
"pad_token": "<pad>",
|
4 |
+
"unk_token": "<unk>"
|
5 |
+
}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae783abfd53d7e55faa057cf0f81b27fd975ae81b114af936b48c3819326d950
|
3 |
+
size 756717
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": null,
|
3 |
+
"clean_up_tokenization_spaces": true,
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"extra_ids": 0,
|
6 |
+
"legacy": false,
|
7 |
+
"model_max_length": 1000000000000000019884624838656,
|
8 |
+
"pad_token": "<pad>",
|
9 |
+
"sp_model_kwargs": {},
|
10 |
+
"tokenizer_class": "T5Tokenizer",
|
11 |
+
"unk_token": "<unk>"
|
12 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,4018 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.634265422821045,
|
3 |
+
"best_model_checkpoint": "./results/checkpoint-5800",
|
4 |
+
"epoch": 0.2142434988179669,
|
5 |
+
"global_step": 5800,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.0,
|
12 |
+
"learning_rate": 0.00019996,
|
13 |
+
"loss": 4.2732,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.0,
|
18 |
+
"learning_rate": 0.00019992000000000002,
|
19 |
+
"loss": 3.4267,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.0,
|
24 |
+
"learning_rate": 0.00019988,
|
25 |
+
"loss": 3.1308,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.0,
|
30 |
+
"learning_rate": 0.00019984,
|
31 |
+
"loss": 2.9584,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.0,
|
36 |
+
"learning_rate": 0.0001998,
|
37 |
+
"loss": 2.8321,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.0,
|
42 |
+
"learning_rate": 0.00019976000000000003,
|
43 |
+
"loss": 2.8353,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.0,
|
48 |
+
"learning_rate": 0.00019972000000000002,
|
49 |
+
"loss": 2.7275,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.0,
|
54 |
+
"learning_rate": 0.00019968,
|
55 |
+
"loss": 2.7678,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.0,
|
60 |
+
"learning_rate": 0.00019964,
|
61 |
+
"loss": 2.6752,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.0,
|
66 |
+
"learning_rate": 0.0001996,
|
67 |
+
"loss": 2.7207,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.0,
|
72 |
+
"eval_accuracy": 0.6036520746718839,
|
73 |
+
"eval_loss": 2.011291980743408,
|
74 |
+
"eval_runtime": 48.3273,
|
75 |
+
"eval_samples_per_second": 8.753,
|
76 |
+
"eval_steps_per_second": 1.097,
|
77 |
+
"step": 100
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 0.0,
|
81 |
+
"learning_rate": 0.00019956000000000002,
|
82 |
+
"loss": 2.6655,
|
83 |
+
"step": 110
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 0.0,
|
87 |
+
"learning_rate": 0.00019952000000000001,
|
88 |
+
"loss": 2.5982,
|
89 |
+
"step": 120
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.0,
|
93 |
+
"learning_rate": 0.00019948,
|
94 |
+
"loss": 2.5936,
|
95 |
+
"step": 130
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.01,
|
99 |
+
"learning_rate": 0.00019944,
|
100 |
+
"loss": 2.625,
|
101 |
+
"step": 140
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.01,
|
105 |
+
"learning_rate": 0.00019940000000000002,
|
106 |
+
"loss": 2.6117,
|
107 |
+
"step": 150
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.01,
|
111 |
+
"learning_rate": 0.00019936000000000002,
|
112 |
+
"loss": 2.5167,
|
113 |
+
"step": 160
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 0.01,
|
117 |
+
"learning_rate": 0.00019932,
|
118 |
+
"loss": 2.4674,
|
119 |
+
"step": 170
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.01,
|
123 |
+
"learning_rate": 0.00019928,
|
124 |
+
"loss": 2.5246,
|
125 |
+
"step": 180
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 0.01,
|
129 |
+
"learning_rate": 0.00019924,
|
130 |
+
"loss": 2.5045,
|
131 |
+
"step": 190
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 0.01,
|
135 |
+
"learning_rate": 0.00019920000000000002,
|
136 |
+
"loss": 2.5222,
|
137 |
+
"step": 200
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.01,
|
141 |
+
"eval_accuracy": 0.5987609032363251,
|
142 |
+
"eval_loss": 2.0005741119384766,
|
143 |
+
"eval_runtime": 47.4839,
|
144 |
+
"eval_samples_per_second": 8.908,
|
145 |
+
"eval_steps_per_second": 1.116,
|
146 |
+
"step": 200
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.01,
|
150 |
+
"learning_rate": 0.00019916,
|
151 |
+
"loss": 2.5048,
|
152 |
+
"step": 210
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.01,
|
156 |
+
"learning_rate": 0.00019912,
|
157 |
+
"loss": 2.4859,
|
158 |
+
"step": 220
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.01,
|
162 |
+
"learning_rate": 0.00019908,
|
163 |
+
"loss": 2.439,
|
164 |
+
"step": 230
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.01,
|
168 |
+
"learning_rate": 0.00019904,
|
169 |
+
"loss": 2.489,
|
170 |
+
"step": 240
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.01,
|
174 |
+
"learning_rate": 0.000199,
|
175 |
+
"loss": 2.5181,
|
176 |
+
"step": 250
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.01,
|
180 |
+
"learning_rate": 0.00019896,
|
181 |
+
"loss": 2.4656,
|
182 |
+
"step": 260
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.01,
|
186 |
+
"learning_rate": 0.00019892000000000003,
|
187 |
+
"loss": 2.4849,
|
188 |
+
"step": 270
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.01,
|
192 |
+
"learning_rate": 0.00019888,
|
193 |
+
"loss": 2.4934,
|
194 |
+
"step": 280
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.01,
|
198 |
+
"learning_rate": 0.00019884000000000001,
|
199 |
+
"loss": 2.4359,
|
200 |
+
"step": 290
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.01,
|
204 |
+
"learning_rate": 0.0001988,
|
205 |
+
"loss": 2.4123,
|
206 |
+
"step": 300
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.01,
|
210 |
+
"eval_accuracy": 0.6184886280264124,
|
211 |
+
"eval_loss": 1.9406265020370483,
|
212 |
+
"eval_runtime": 50.1664,
|
213 |
+
"eval_samples_per_second": 8.432,
|
214 |
+
"eval_steps_per_second": 1.056,
|
215 |
+
"step": 300
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 0.01,
|
219 |
+
"learning_rate": 0.00019876,
|
220 |
+
"loss": 2.4585,
|
221 |
+
"step": 310
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.01,
|
225 |
+
"learning_rate": 0.00019872000000000002,
|
226 |
+
"loss": 2.436,
|
227 |
+
"step": 320
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.01,
|
231 |
+
"learning_rate": 0.00019868,
|
232 |
+
"loss": 2.4361,
|
233 |
+
"step": 330
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.01,
|
237 |
+
"learning_rate": 0.00019864,
|
238 |
+
"loss": 2.4193,
|
239 |
+
"step": 340
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.01,
|
243 |
+
"learning_rate": 0.0001986,
|
244 |
+
"loss": 2.4404,
|
245 |
+
"step": 350
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 0.01,
|
249 |
+
"learning_rate": 0.00019856000000000002,
|
250 |
+
"loss": 2.4109,
|
251 |
+
"step": 360
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 0.01,
|
255 |
+
"learning_rate": 0.00019852000000000002,
|
256 |
+
"loss": 2.4257,
|
257 |
+
"step": 370
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.01,
|
261 |
+
"learning_rate": 0.00019848,
|
262 |
+
"loss": 2.423,
|
263 |
+
"step": 380
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 0.01,
|
267 |
+
"learning_rate": 0.00019844,
|
268 |
+
"loss": 2.3701,
|
269 |
+
"step": 390
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.01,
|
273 |
+
"learning_rate": 0.0001984,
|
274 |
+
"loss": 2.4469,
|
275 |
+
"step": 400
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.01,
|
279 |
+
"eval_accuracy": 0.6204450966006358,
|
280 |
+
"eval_loss": 1.9681750535964966,
|
281 |
+
"eval_runtime": 47.2918,
|
282 |
+
"eval_samples_per_second": 8.944,
|
283 |
+
"eval_steps_per_second": 1.121,
|
284 |
+
"step": 400
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.02,
|
288 |
+
"learning_rate": 0.00019836000000000002,
|
289 |
+
"loss": 2.3953,
|
290 |
+
"step": 410
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.02,
|
294 |
+
"learning_rate": 0.00019832,
|
295 |
+
"loss": 2.3848,
|
296 |
+
"step": 420
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.02,
|
300 |
+
"learning_rate": 0.00019828,
|
301 |
+
"loss": 2.3802,
|
302 |
+
"step": 430
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.02,
|
306 |
+
"learning_rate": 0.00019824,
|
307 |
+
"loss": 2.4174,
|
308 |
+
"step": 440
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.02,
|
312 |
+
"learning_rate": 0.00019820000000000002,
|
313 |
+
"loss": 2.4125,
|
314 |
+
"step": 450
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.02,
|
318 |
+
"learning_rate": 0.00019816000000000001,
|
319 |
+
"loss": 2.387,
|
320 |
+
"step": 460
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.02,
|
324 |
+
"learning_rate": 0.00019812,
|
325 |
+
"loss": 2.3491,
|
326 |
+
"step": 470
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.02,
|
330 |
+
"learning_rate": 0.00019808,
|
331 |
+
"loss": 2.3997,
|
332 |
+
"step": 480
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.02,
|
336 |
+
"learning_rate": 0.00019804,
|
337 |
+
"loss": 2.3107,
|
338 |
+
"step": 490
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.02,
|
342 |
+
"learning_rate": 0.00019800000000000002,
|
343 |
+
"loss": 2.3367,
|
344 |
+
"step": 500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.02,
|
348 |
+
"eval_accuracy": 0.6377272356729436,
|
349 |
+
"eval_loss": 1.8571785688400269,
|
350 |
+
"eval_runtime": 47.2012,
|
351 |
+
"eval_samples_per_second": 8.962,
|
352 |
+
"eval_steps_per_second": 1.123,
|
353 |
+
"step": 500
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 0.02,
|
357 |
+
"learning_rate": 0.00019796,
|
358 |
+
"loss": 2.3571,
|
359 |
+
"step": 510
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.02,
|
363 |
+
"learning_rate": 0.00019792000000000003,
|
364 |
+
"loss": 2.325,
|
365 |
+
"step": 520
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 0.02,
|
369 |
+
"learning_rate": 0.00019788,
|
370 |
+
"loss": 2.39,
|
371 |
+
"step": 530
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 0.02,
|
375 |
+
"learning_rate": 0.00019784,
|
376 |
+
"loss": 2.3508,
|
377 |
+
"step": 540
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 0.02,
|
381 |
+
"learning_rate": 0.0001978,
|
382 |
+
"loss": 2.3736,
|
383 |
+
"step": 550
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 0.02,
|
387 |
+
"learning_rate": 0.00019776,
|
388 |
+
"loss": 2.2891,
|
389 |
+
"step": 560
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.02,
|
393 |
+
"learning_rate": 0.00019772000000000002,
|
394 |
+
"loss": 2.3026,
|
395 |
+
"step": 570
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 0.02,
|
399 |
+
"learning_rate": 0.00019768,
|
400 |
+
"loss": 2.2965,
|
401 |
+
"step": 580
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.02,
|
405 |
+
"learning_rate": 0.00019764,
|
406 |
+
"loss": 2.3467,
|
407 |
+
"step": 590
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 0.02,
|
411 |
+
"learning_rate": 0.0001976,
|
412 |
+
"loss": 2.2691,
|
413 |
+
"step": 600
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 0.02,
|
417 |
+
"eval_accuracy": 0.6315317518545692,
|
418 |
+
"eval_loss": 1.8730525970458984,
|
419 |
+
"eval_runtime": 46.6226,
|
420 |
+
"eval_samples_per_second": 9.073,
|
421 |
+
"eval_steps_per_second": 1.137,
|
422 |
+
"step": 600
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.02,
|
426 |
+
"learning_rate": 0.00019756,
|
427 |
+
"loss": 2.3046,
|
428 |
+
"step": 610
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.02,
|
432 |
+
"learning_rate": 0.00019752000000000002,
|
433 |
+
"loss": 2.2575,
|
434 |
+
"step": 620
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.02,
|
438 |
+
"learning_rate": 0.00019748,
|
439 |
+
"loss": 2.2895,
|
440 |
+
"step": 630
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.02,
|
444 |
+
"learning_rate": 0.00019744,
|
445 |
+
"loss": 2.2699,
|
446 |
+
"step": 640
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.02,
|
450 |
+
"learning_rate": 0.0001974,
|
451 |
+
"loss": 2.2235,
|
452 |
+
"step": 650
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.02,
|
456 |
+
"learning_rate": 0.00019736000000000002,
|
457 |
+
"loss": 2.2322,
|
458 |
+
"step": 660
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.02,
|
462 |
+
"learning_rate": 0.00019732000000000001,
|
463 |
+
"loss": 2.2743,
|
464 |
+
"step": 670
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.03,
|
468 |
+
"learning_rate": 0.00019728,
|
469 |
+
"loss": 2.2625,
|
470 |
+
"step": 680
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.03,
|
474 |
+
"learning_rate": 0.00019724,
|
475 |
+
"loss": 2.2843,
|
476 |
+
"step": 690
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.03,
|
480 |
+
"learning_rate": 0.0001972,
|
481 |
+
"loss": 2.2715,
|
482 |
+
"step": 700
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 0.03,
|
486 |
+
"eval_accuracy": 0.632428466617755,
|
487 |
+
"eval_loss": 1.8891419172286987,
|
488 |
+
"eval_runtime": 47.1404,
|
489 |
+
"eval_samples_per_second": 8.973,
|
490 |
+
"eval_steps_per_second": 1.124,
|
491 |
+
"step": 700
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 0.03,
|
495 |
+
"learning_rate": 0.00019716000000000002,
|
496 |
+
"loss": 2.2755,
|
497 |
+
"step": 710
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 0.03,
|
501 |
+
"learning_rate": 0.00019712,
|
502 |
+
"loss": 2.2597,
|
503 |
+
"step": 720
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 0.03,
|
507 |
+
"learning_rate": 0.00019708000000000003,
|
508 |
+
"loss": 2.2185,
|
509 |
+
"step": 730
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 0.03,
|
513 |
+
"learning_rate": 0.00019704,
|
514 |
+
"loss": 2.3046,
|
515 |
+
"step": 740
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 0.03,
|
519 |
+
"learning_rate": 0.00019700000000000002,
|
520 |
+
"loss": 2.3451,
|
521 |
+
"step": 750
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 0.03,
|
525 |
+
"learning_rate": 0.00019696,
|
526 |
+
"loss": 2.2956,
|
527 |
+
"step": 760
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.03,
|
531 |
+
"learning_rate": 0.00019692,
|
532 |
+
"loss": 2.2371,
|
533 |
+
"step": 770
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 0.03,
|
537 |
+
"learning_rate": 0.00019688000000000003,
|
538 |
+
"loss": 2.2705,
|
539 |
+
"step": 780
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 0.03,
|
543 |
+
"learning_rate": 0.00019684,
|
544 |
+
"loss": 2.2141,
|
545 |
+
"step": 790
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 0.03,
|
549 |
+
"learning_rate": 0.0001968,
|
550 |
+
"loss": 2.2344,
|
551 |
+
"step": 800
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 0.03,
|
555 |
+
"eval_accuracy": 0.6379717942447216,
|
556 |
+
"eval_loss": 1.8463128805160522,
|
557 |
+
"eval_runtime": 47.4067,
|
558 |
+
"eval_samples_per_second": 8.923,
|
559 |
+
"eval_steps_per_second": 1.118,
|
560 |
+
"step": 800
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 0.03,
|
564 |
+
"learning_rate": 0.00019676,
|
565 |
+
"loss": 2.221,
|
566 |
+
"step": 810
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 0.03,
|
570 |
+
"learning_rate": 0.00019672000000000003,
|
571 |
+
"loss": 2.3338,
|
572 |
+
"step": 820
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 0.03,
|
576 |
+
"learning_rate": 0.00019668000000000002,
|
577 |
+
"loss": 2.1978,
|
578 |
+
"step": 830
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 0.03,
|
582 |
+
"learning_rate": 0.00019664000000000001,
|
583 |
+
"loss": 2.2532,
|
584 |
+
"step": 840
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 0.03,
|
588 |
+
"learning_rate": 0.0001966,
|
589 |
+
"loss": 2.2765,
|
590 |
+
"step": 850
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.03,
|
594 |
+
"learning_rate": 0.00019656,
|
595 |
+
"loss": 2.2643,
|
596 |
+
"step": 860
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 0.03,
|
600 |
+
"learning_rate": 0.00019652000000000002,
|
601 |
+
"loss": 2.2794,
|
602 |
+
"step": 870
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 0.03,
|
606 |
+
"learning_rate": 0.00019648000000000002,
|
607 |
+
"loss": 2.2269,
|
608 |
+
"step": 880
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 0.03,
|
612 |
+
"learning_rate": 0.00019644,
|
613 |
+
"loss": 2.2576,
|
614 |
+
"step": 890
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 0.03,
|
618 |
+
"learning_rate": 0.0001964,
|
619 |
+
"loss": 2.2234,
|
620 |
+
"step": 900
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 0.03,
|
624 |
+
"eval_accuracy": 0.632183908045977,
|
625 |
+
"eval_loss": 1.866058588027954,
|
626 |
+
"eval_runtime": 46.4721,
|
627 |
+
"eval_samples_per_second": 9.102,
|
628 |
+
"eval_steps_per_second": 1.14,
|
629 |
+
"step": 900
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 0.03,
|
633 |
+
"learning_rate": 0.00019636000000000002,
|
634 |
+
"loss": 2.2282,
|
635 |
+
"step": 910
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 0.03,
|
639 |
+
"learning_rate": 0.00019632000000000002,
|
640 |
+
"loss": 2.2236,
|
641 |
+
"step": 920
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 0.03,
|
645 |
+
"learning_rate": 0.00019628,
|
646 |
+
"loss": 2.2008,
|
647 |
+
"step": 930
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 0.03,
|
651 |
+
"learning_rate": 0.00019624,
|
652 |
+
"loss": 2.2731,
|
653 |
+
"step": 940
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.04,
|
657 |
+
"learning_rate": 0.0001962,
|
658 |
+
"loss": 2.2424,
|
659 |
+
"step": 950
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 0.04,
|
663 |
+
"learning_rate": 0.00019616000000000002,
|
664 |
+
"loss": 2.1774,
|
665 |
+
"step": 960
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 0.04,
|
669 |
+
"learning_rate": 0.00019612,
|
670 |
+
"loss": 2.1652,
|
671 |
+
"step": 970
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 0.04,
|
675 |
+
"learning_rate": 0.00019608,
|
676 |
+
"loss": 2.1703,
|
677 |
+
"step": 980
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 0.04,
|
681 |
+
"learning_rate": 0.00019604,
|
682 |
+
"loss": 2.185,
|
683 |
+
"step": 990
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 0.04,
|
687 |
+
"learning_rate": 0.000196,
|
688 |
+
"loss": 2.1818,
|
689 |
+
"step": 1000
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.04,
|
693 |
+
"eval_accuracy": 0.6382163528164996,
|
694 |
+
"eval_loss": 1.81806218624115,
|
695 |
+
"eval_runtime": 60.7049,
|
696 |
+
"eval_samples_per_second": 6.968,
|
697 |
+
"eval_steps_per_second": 0.873,
|
698 |
+
"step": 1000
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 0.04,
|
702 |
+
"learning_rate": 0.00019596000000000001,
|
703 |
+
"loss": 2.1961,
|
704 |
+
"step": 1010
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 0.04,
|
708 |
+
"learning_rate": 0.00019592,
|
709 |
+
"loss": 2.1806,
|
710 |
+
"step": 1020
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 0.04,
|
714 |
+
"learning_rate": 0.00019588000000000003,
|
715 |
+
"loss": 2.1523,
|
716 |
+
"step": 1030
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.04,
|
720 |
+
"learning_rate": 0.00019584,
|
721 |
+
"loss": 2.1305,
|
722 |
+
"step": 1040
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 0.04,
|
726 |
+
"learning_rate": 0.00019580000000000002,
|
727 |
+
"loss": 2.1467,
|
728 |
+
"step": 1050
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 0.04,
|
732 |
+
"learning_rate": 0.00019576,
|
733 |
+
"loss": 2.1324,
|
734 |
+
"step": 1060
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 0.04,
|
738 |
+
"learning_rate": 0.00019572,
|
739 |
+
"loss": 2.211,
|
740 |
+
"step": 1070
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 0.04,
|
744 |
+
"learning_rate": 0.00019568000000000002,
|
745 |
+
"loss": 2.1626,
|
746 |
+
"step": 1080
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 0.04,
|
750 |
+
"learning_rate": 0.00019564,
|
751 |
+
"loss": 2.1327,
|
752 |
+
"step": 1090
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 0.04,
|
756 |
+
"learning_rate": 0.0001956,
|
757 |
+
"loss": 2.1808,
|
758 |
+
"step": 1100
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.04,
|
762 |
+
"eval_accuracy": 0.6419662509170947,
|
763 |
+
"eval_loss": 1.838297724723816,
|
764 |
+
"eval_runtime": 61.0821,
|
765 |
+
"eval_samples_per_second": 6.925,
|
766 |
+
"eval_steps_per_second": 0.868,
|
767 |
+
"step": 1100
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 0.04,
|
771 |
+
"learning_rate": 0.00019556,
|
772 |
+
"loss": 2.2277,
|
773 |
+
"step": 1110
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 0.04,
|
777 |
+
"learning_rate": 0.00019552000000000003,
|
778 |
+
"loss": 2.1227,
|
779 |
+
"step": 1120
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.04,
|
783 |
+
"learning_rate": 0.00019548000000000002,
|
784 |
+
"loss": 2.2336,
|
785 |
+
"step": 1130
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 0.04,
|
789 |
+
"learning_rate": 0.00019544,
|
790 |
+
"loss": 2.1708,
|
791 |
+
"step": 1140
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 0.04,
|
795 |
+
"learning_rate": 0.0001954,
|
796 |
+
"loss": 2.1549,
|
797 |
+
"step": 1150
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 0.04,
|
801 |
+
"learning_rate": 0.00019536,
|
802 |
+
"loss": 2.1776,
|
803 |
+
"step": 1160
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 0.04,
|
807 |
+
"learning_rate": 0.00019532000000000002,
|
808 |
+
"loss": 2.1634,
|
809 |
+
"step": 1170
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 0.04,
|
813 |
+
"learning_rate": 0.00019528000000000001,
|
814 |
+
"loss": 2.1891,
|
815 |
+
"step": 1180
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 0.04,
|
819 |
+
"learning_rate": 0.00019524,
|
820 |
+
"loss": 2.2221,
|
821 |
+
"step": 1190
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.04,
|
825 |
+
"learning_rate": 0.0001952,
|
826 |
+
"loss": 2.1885,
|
827 |
+
"step": 1200
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 0.04,
|
831 |
+
"eval_accuracy": 0.6440857585391702,
|
832 |
+
"eval_loss": 1.8327720165252686,
|
833 |
+
"eval_runtime": 61.1648,
|
834 |
+
"eval_samples_per_second": 6.916,
|
835 |
+
"eval_steps_per_second": 0.867,
|
836 |
+
"step": 1200
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.04,
|
840 |
+
"learning_rate": 0.00019516000000000002,
|
841 |
+
"loss": 2.1854,
|
842 |
+
"step": 1210
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.05,
|
846 |
+
"learning_rate": 0.00019512000000000002,
|
847 |
+
"loss": 2.1554,
|
848 |
+
"step": 1220
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 0.05,
|
852 |
+
"learning_rate": 0.00019508,
|
853 |
+
"loss": 2.1365,
|
854 |
+
"step": 1230
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 0.05,
|
858 |
+
"learning_rate": 0.00019504,
|
859 |
+
"loss": 2.1783,
|
860 |
+
"step": 1240
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 0.05,
|
864 |
+
"learning_rate": 0.000195,
|
865 |
+
"loss": 2.1918,
|
866 |
+
"step": 1250
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 0.05,
|
870 |
+
"learning_rate": 0.00019496000000000002,
|
871 |
+
"loss": 2.2186,
|
872 |
+
"step": 1260
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 0.05,
|
876 |
+
"learning_rate": 0.00019492,
|
877 |
+
"loss": 2.1508,
|
878 |
+
"step": 1270
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 0.05,
|
882 |
+
"learning_rate": 0.00019488000000000003,
|
883 |
+
"loss": 2.1997,
|
884 |
+
"step": 1280
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.05,
|
888 |
+
"learning_rate": 0.00019484,
|
889 |
+
"loss": 2.0978,
|
890 |
+
"step": 1290
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 0.05,
|
894 |
+
"learning_rate": 0.0001948,
|
895 |
+
"loss": 2.1547,
|
896 |
+
"step": 1300
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 0.05,
|
900 |
+
"eval_accuracy": 0.648650851879025,
|
901 |
+
"eval_loss": 1.7967556715011597,
|
902 |
+
"eval_runtime": 60.331,
|
903 |
+
"eval_samples_per_second": 7.011,
|
904 |
+
"eval_steps_per_second": 0.878,
|
905 |
+
"step": 1300
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.05,
|
909 |
+
"learning_rate": 0.00019476,
|
910 |
+
"loss": 2.1202,
|
911 |
+
"step": 1310
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 0.05,
|
915 |
+
"learning_rate": 0.00019472,
|
916 |
+
"loss": 2.1406,
|
917 |
+
"step": 1320
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 0.05,
|
921 |
+
"learning_rate": 0.00019468000000000003,
|
922 |
+
"loss": 2.149,
|
923 |
+
"step": 1330
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 0.05,
|
927 |
+
"learning_rate": 0.00019464,
|
928 |
+
"loss": 2.1915,
|
929 |
+
"step": 1340
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 0.05,
|
933 |
+
"learning_rate": 0.00019460000000000001,
|
934 |
+
"loss": 2.1721,
|
935 |
+
"step": 1350
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 0.05,
|
939 |
+
"learning_rate": 0.00019456,
|
940 |
+
"loss": 2.1155,
|
941 |
+
"step": 1360
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 0.05,
|
945 |
+
"learning_rate": 0.00019452,
|
946 |
+
"loss": 2.138,
|
947 |
+
"step": 1370
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.05,
|
951 |
+
"learning_rate": 0.00019448000000000002,
|
952 |
+
"loss": 2.0946,
|
953 |
+
"step": 1380
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 0.05,
|
957 |
+
"learning_rate": 0.00019444,
|
958 |
+
"loss": 2.1564,
|
959 |
+
"step": 1390
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 0.05,
|
963 |
+
"learning_rate": 0.0001944,
|
964 |
+
"loss": 2.1261,
|
965 |
+
"step": 1400
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 0.05,
|
969 |
+
"eval_accuracy": 0.6471835004483574,
|
970 |
+
"eval_loss": 1.8035624027252197,
|
971 |
+
"eval_runtime": 59.6014,
|
972 |
+
"eval_samples_per_second": 7.097,
|
973 |
+
"eval_steps_per_second": 0.889,
|
974 |
+
"step": 1400
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 0.05,
|
978 |
+
"learning_rate": 0.00019436,
|
979 |
+
"loss": 2.1554,
|
980 |
+
"step": 1410
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 0.05,
|
984 |
+
"learning_rate": 0.00019432000000000002,
|
985 |
+
"loss": 2.1857,
|
986 |
+
"step": 1420
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 0.05,
|
990 |
+
"learning_rate": 0.00019428000000000002,
|
991 |
+
"loss": 2.1381,
|
992 |
+
"step": 1430
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 0.05,
|
996 |
+
"learning_rate": 0.00019424,
|
997 |
+
"loss": 2.1638,
|
998 |
+
"step": 1440
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 0.05,
|
1002 |
+
"learning_rate": 0.0001942,
|
1003 |
+
"loss": 2.1491,
|
1004 |
+
"step": 1450
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 0.05,
|
1008 |
+
"learning_rate": 0.00019416,
|
1009 |
+
"loss": 2.209,
|
1010 |
+
"step": 1460
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.05,
|
1014 |
+
"learning_rate": 0.00019412000000000002,
|
1015 |
+
"loss": 2.2035,
|
1016 |
+
"step": 1470
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 0.05,
|
1020 |
+
"learning_rate": 0.00019408,
|
1021 |
+
"loss": 2.079,
|
1022 |
+
"step": 1480
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 0.06,
|
1026 |
+
"learning_rate": 0.00019404,
|
1027 |
+
"loss": 2.1535,
|
1028 |
+
"step": 1490
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 0.06,
|
1032 |
+
"learning_rate": 0.000194,
|
1033 |
+
"loss": 2.1074,
|
1034 |
+
"step": 1500
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 0.06,
|
1038 |
+
"eval_accuracy": 0.6522377109317682,
|
1039 |
+
"eval_loss": 1.7710509300231934,
|
1040 |
+
"eval_runtime": 45.8627,
|
1041 |
+
"eval_samples_per_second": 9.223,
|
1042 |
+
"eval_steps_per_second": 1.156,
|
1043 |
+
"step": 1500
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 0.06,
|
1047 |
+
"learning_rate": 0.00019396000000000002,
|
1048 |
+
"loss": 2.0692,
|
1049 |
+
"step": 1510
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 0.06,
|
1053 |
+
"learning_rate": 0.00019392000000000001,
|
1054 |
+
"loss": 2.106,
|
1055 |
+
"step": 1520
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 0.06,
|
1059 |
+
"learning_rate": 0.00019388,
|
1060 |
+
"loss": 2.133,
|
1061 |
+
"step": 1530
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 0.06,
|
1065 |
+
"learning_rate": 0.00019384,
|
1066 |
+
"loss": 2.0844,
|
1067 |
+
"step": 1540
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 0.06,
|
1071 |
+
"learning_rate": 0.0001938,
|
1072 |
+
"loss": 2.0839,
|
1073 |
+
"step": 1550
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.06,
|
1077 |
+
"learning_rate": 0.00019376000000000002,
|
1078 |
+
"loss": 2.1023,
|
1079 |
+
"step": 1560
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 0.06,
|
1083 |
+
"learning_rate": 0.00019372,
|
1084 |
+
"loss": 2.124,
|
1085 |
+
"step": 1570
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 0.06,
|
1089 |
+
"learning_rate": 0.00019368000000000003,
|
1090 |
+
"loss": 2.1142,
|
1091 |
+
"step": 1580
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 0.06,
|
1095 |
+
"learning_rate": 0.00019364,
|
1096 |
+
"loss": 2.0142,
|
1097 |
+
"step": 1590
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 0.06,
|
1101 |
+
"learning_rate": 0.00019360000000000002,
|
1102 |
+
"loss": 2.1537,
|
1103 |
+
"step": 1600
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 0.06,
|
1107 |
+
"eval_accuracy": 0.6504442814053966,
|
1108 |
+
"eval_loss": 1.7606570720672607,
|
1109 |
+
"eval_runtime": 46.4971,
|
1110 |
+
"eval_samples_per_second": 9.097,
|
1111 |
+
"eval_steps_per_second": 1.14,
|
1112 |
+
"step": 1600
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 0.06,
|
1116 |
+
"learning_rate": 0.00019356,
|
1117 |
+
"loss": 2.0624,
|
1118 |
+
"step": 1610
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 0.06,
|
1122 |
+
"learning_rate": 0.00019352,
|
1123 |
+
"loss": 2.1308,
|
1124 |
+
"step": 1620
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 0.06,
|
1128 |
+
"learning_rate": 0.00019348000000000002,
|
1129 |
+
"loss": 2.0587,
|
1130 |
+
"step": 1630
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 0.06,
|
1134 |
+
"learning_rate": 0.00019344,
|
1135 |
+
"loss": 2.1031,
|
1136 |
+
"step": 1640
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.06,
|
1140 |
+
"learning_rate": 0.0001934,
|
1141 |
+
"loss": 2.0765,
|
1142 |
+
"step": 1650
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 0.06,
|
1146 |
+
"learning_rate": 0.00019336,
|
1147 |
+
"loss": 2.0972,
|
1148 |
+
"step": 1660
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 0.06,
|
1152 |
+
"learning_rate": 0.00019332,
|
1153 |
+
"loss": 2.0875,
|
1154 |
+
"step": 1670
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 0.06,
|
1158 |
+
"learning_rate": 0.00019328000000000002,
|
1159 |
+
"loss": 2.1277,
|
1160 |
+
"step": 1680
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 0.06,
|
1164 |
+
"learning_rate": 0.00019323999999999999,
|
1165 |
+
"loss": 2.088,
|
1166 |
+
"step": 1690
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 0.06,
|
1170 |
+
"learning_rate": 0.0001932,
|
1171 |
+
"loss": 2.1085,
|
1172 |
+
"step": 1700
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 0.06,
|
1176 |
+
"eval_accuracy": 0.6466943833048014,
|
1177 |
+
"eval_loss": 1.7533295154571533,
|
1178 |
+
"eval_runtime": 45.9496,
|
1179 |
+
"eval_samples_per_second": 9.206,
|
1180 |
+
"eval_steps_per_second": 1.153,
|
1181 |
+
"step": 1700
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 0.06,
|
1185 |
+
"learning_rate": 0.00019316,
|
1186 |
+
"loss": 2.159,
|
1187 |
+
"step": 1710
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 0.06,
|
1191 |
+
"learning_rate": 0.00019312000000000002,
|
1192 |
+
"loss": 2.0808,
|
1193 |
+
"step": 1720
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 0.06,
|
1197 |
+
"learning_rate": 0.00019308000000000001,
|
1198 |
+
"loss": 2.1481,
|
1199 |
+
"step": 1730
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.06,
|
1203 |
+
"learning_rate": 0.00019304,
|
1204 |
+
"loss": 2.0853,
|
1205 |
+
"step": 1740
|
1206 |
+
},
|
1207 |
+
{
|
1208 |
+
"epoch": 0.06,
|
1209 |
+
"learning_rate": 0.000193,
|
1210 |
+
"loss": 2.0692,
|
1211 |
+
"step": 1750
|
1212 |
+
},
|
1213 |
+
{
|
1214 |
+
"epoch": 0.07,
|
1215 |
+
"learning_rate": 0.00019296,
|
1216 |
+
"loss": 2.2094,
|
1217 |
+
"step": 1760
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 0.07,
|
1221 |
+
"learning_rate": 0.00019292000000000002,
|
1222 |
+
"loss": 2.0793,
|
1223 |
+
"step": 1770
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 0.07,
|
1227 |
+
"learning_rate": 0.00019288,
|
1228 |
+
"loss": 2.0469,
|
1229 |
+
"step": 1780
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 0.07,
|
1233 |
+
"learning_rate": 0.00019284,
|
1234 |
+
"loss": 2.1105,
|
1235 |
+
"step": 1790
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 0.07,
|
1239 |
+
"learning_rate": 0.0001928,
|
1240 |
+
"loss": 2.1268,
|
1241 |
+
"step": 1800
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.07,
|
1245 |
+
"eval_accuracy": 0.6550093747452514,
|
1246 |
+
"eval_loss": 1.7450594902038574,
|
1247 |
+
"eval_runtime": 46.1382,
|
1248 |
+
"eval_samples_per_second": 9.168,
|
1249 |
+
"eval_steps_per_second": 1.149,
|
1250 |
+
"step": 1800
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 0.07,
|
1254 |
+
"learning_rate": 0.00019276000000000002,
|
1255 |
+
"loss": 2.0895,
|
1256 |
+
"step": 1810
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 0.07,
|
1260 |
+
"learning_rate": 0.00019272,
|
1261 |
+
"loss": 2.1127,
|
1262 |
+
"step": 1820
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.07,
|
1266 |
+
"learning_rate": 0.00019268,
|
1267 |
+
"loss": 2.0435,
|
1268 |
+
"step": 1830
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 0.07,
|
1272 |
+
"learning_rate": 0.00019264,
|
1273 |
+
"loss": 2.0799,
|
1274 |
+
"step": 1840
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 0.07,
|
1278 |
+
"learning_rate": 0.0001926,
|
1279 |
+
"loss": 2.0886,
|
1280 |
+
"step": 1850
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 0.07,
|
1284 |
+
"learning_rate": 0.00019256,
|
1285 |
+
"loss": 2.0726,
|
1286 |
+
"step": 1860
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 0.07,
|
1290 |
+
"learning_rate": 0.00019252,
|
1291 |
+
"loss": 2.1095,
|
1292 |
+
"step": 1870
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 0.07,
|
1296 |
+
"learning_rate": 0.00019248000000000003,
|
1297 |
+
"loss": 2.0732,
|
1298 |
+
"step": 1880
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 0.07,
|
1302 |
+
"learning_rate": 0.00019244000000000002,
|
1303 |
+
"loss": 2.0577,
|
1304 |
+
"step": 1890
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.07,
|
1308 |
+
"learning_rate": 0.00019240000000000001,
|
1309 |
+
"loss": 2.0991,
|
1310 |
+
"step": 1900
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 0.07,
|
1314 |
+
"eval_accuracy": 0.6504442814053966,
|
1315 |
+
"eval_loss": 1.7744449377059937,
|
1316 |
+
"eval_runtime": 46.1319,
|
1317 |
+
"eval_samples_per_second": 9.169,
|
1318 |
+
"eval_steps_per_second": 1.149,
|
1319 |
+
"step": 1900
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 0.07,
|
1323 |
+
"learning_rate": 0.00019236,
|
1324 |
+
"loss": 2.113,
|
1325 |
+
"step": 1910
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.07,
|
1329 |
+
"learning_rate": 0.00019232,
|
1330 |
+
"loss": 2.0054,
|
1331 |
+
"step": 1920
|
1332 |
+
},
|
1333 |
+
{
|
1334 |
+
"epoch": 0.07,
|
1335 |
+
"learning_rate": 0.00019228000000000002,
|
1336 |
+
"loss": 2.0531,
|
1337 |
+
"step": 1930
|
1338 |
+
},
|
1339 |
+
{
|
1340 |
+
"epoch": 0.07,
|
1341 |
+
"learning_rate": 0.00019224000000000002,
|
1342 |
+
"loss": 2.0268,
|
1343 |
+
"step": 1940
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 0.07,
|
1347 |
+
"learning_rate": 0.0001922,
|
1348 |
+
"loss": 2.0703,
|
1349 |
+
"step": 1950
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 0.07,
|
1353 |
+
"learning_rate": 0.00019216,
|
1354 |
+
"loss": 2.0101,
|
1355 |
+
"step": 1960
|
1356 |
+
},
|
1357 |
+
{
|
1358 |
+
"epoch": 0.07,
|
1359 |
+
"learning_rate": 0.00019212000000000002,
|
1360 |
+
"loss": 2.0598,
|
1361 |
+
"step": 1970
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 0.07,
|
1365 |
+
"learning_rate": 0.00019208000000000002,
|
1366 |
+
"loss": 2.0731,
|
1367 |
+
"step": 1980
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.07,
|
1371 |
+
"learning_rate": 0.00019204,
|
1372 |
+
"loss": 2.0695,
|
1373 |
+
"step": 1990
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 0.07,
|
1377 |
+
"learning_rate": 0.000192,
|
1378 |
+
"loss": 2.0982,
|
1379 |
+
"step": 2000
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 0.07,
|
1383 |
+
"eval_accuracy": 0.6515855547403603,
|
1384 |
+
"eval_loss": 1.738739252090454,
|
1385 |
+
"eval_runtime": 45.7921,
|
1386 |
+
"eval_samples_per_second": 9.237,
|
1387 |
+
"eval_steps_per_second": 1.157,
|
1388 |
+
"step": 2000
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.07,
|
1392 |
+
"learning_rate": 0.00019196,
|
1393 |
+
"loss": 2.0565,
|
1394 |
+
"step": 2010
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 0.07,
|
1398 |
+
"learning_rate": 0.00019192000000000002,
|
1399 |
+
"loss": 2.0358,
|
1400 |
+
"step": 2020
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 0.07,
|
1404 |
+
"learning_rate": 0.00019188,
|
1405 |
+
"loss": 2.0407,
|
1406 |
+
"step": 2030
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 0.08,
|
1410 |
+
"learning_rate": 0.00019184,
|
1411 |
+
"loss": 2.0584,
|
1412 |
+
"step": 2040
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 0.08,
|
1416 |
+
"learning_rate": 0.0001918,
|
1417 |
+
"loss": 2.0313,
|
1418 |
+
"step": 2050
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 0.08,
|
1422 |
+
"learning_rate": 0.00019176,
|
1423 |
+
"loss": 2.0852,
|
1424 |
+
"step": 2060
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 0.08,
|
1428 |
+
"learning_rate": 0.00019172000000000001,
|
1429 |
+
"loss": 2.1305,
|
1430 |
+
"step": 2070
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.08,
|
1434 |
+
"learning_rate": 0.00019168,
|
1435 |
+
"loss": 2.8885,
|
1436 |
+
"step": 2080
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 0.08,
|
1440 |
+
"learning_rate": 0.00019164000000000003,
|
1441 |
+
"loss": 5.2587,
|
1442 |
+
"step": 2090
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 0.08,
|
1446 |
+
"learning_rate": 0.0001916,
|
1447 |
+
"loss": 5.5295,
|
1448 |
+
"step": 2100
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 0.08,
|
1452 |
+
"eval_accuracy": 0.31556207711746964,
|
1453 |
+
"eval_loss": 4.5144548416137695,
|
1454 |
+
"eval_runtime": 46.4657,
|
1455 |
+
"eval_samples_per_second": 9.103,
|
1456 |
+
"eval_steps_per_second": 1.141,
|
1457 |
+
"step": 2100
|
1458 |
+
},
|
1459 |
+
{
|
1460 |
+
"epoch": 0.08,
|
1461 |
+
"learning_rate": 0.00019156000000000002,
|
1462 |
+
"loss": 4.2499,
|
1463 |
+
"step": 2110
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 0.08,
|
1467 |
+
"learning_rate": 0.00019152,
|
1468 |
+
"loss": 2.2078,
|
1469 |
+
"step": 2120
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 0.08,
|
1473 |
+
"learning_rate": 0.00019148,
|
1474 |
+
"loss": 2.1916,
|
1475 |
+
"step": 2130
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 0.08,
|
1479 |
+
"learning_rate": 0.00019144000000000002,
|
1480 |
+
"loss": 2.1247,
|
1481 |
+
"step": 2140
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 0.08,
|
1485 |
+
"learning_rate": 0.0001914,
|
1486 |
+
"loss": 2.0517,
|
1487 |
+
"step": 2150
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 0.08,
|
1491 |
+
"learning_rate": 0.00019136,
|
1492 |
+
"loss": 2.1537,
|
1493 |
+
"step": 2160
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.08,
|
1497 |
+
"learning_rate": 0.00019132,
|
1498 |
+
"loss": 2.1611,
|
1499 |
+
"step": 2170
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 0.08,
|
1503 |
+
"learning_rate": 0.00019128000000000003,
|
1504 |
+
"loss": 2.1015,
|
1505 |
+
"step": 2180
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 0.08,
|
1509 |
+
"learning_rate": 0.00019124000000000002,
|
1510 |
+
"loss": 2.191,
|
1511 |
+
"step": 2190
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 0.08,
|
1515 |
+
"learning_rate": 0.0001912,
|
1516 |
+
"loss": 2.0354,
|
1517 |
+
"step": 2200
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 0.08,
|
1521 |
+
"eval_accuracy": 0.652645308551398,
|
1522 |
+
"eval_loss": 1.74555242061615,
|
1523 |
+
"eval_runtime": 45.8199,
|
1524 |
+
"eval_samples_per_second": 9.232,
|
1525 |
+
"eval_steps_per_second": 1.157,
|
1526 |
+
"step": 2200
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 0.08,
|
1530 |
+
"learning_rate": 0.00019116,
|
1531 |
+
"loss": 2.0853,
|
1532 |
+
"step": 2210
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 0.08,
|
1536 |
+
"learning_rate": 0.00019112,
|
1537 |
+
"loss": 2.0091,
|
1538 |
+
"step": 2220
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 0.08,
|
1542 |
+
"learning_rate": 0.00019108000000000002,
|
1543 |
+
"loss": 2.093,
|
1544 |
+
"step": 2230
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 0.08,
|
1548 |
+
"learning_rate": 0.00019104000000000001,
|
1549 |
+
"loss": 1.9993,
|
1550 |
+
"step": 2240
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 0.08,
|
1554 |
+
"learning_rate": 0.000191,
|
1555 |
+
"loss": 2.0933,
|
1556 |
+
"step": 2250
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.08,
|
1560 |
+
"learning_rate": 0.00019096,
|
1561 |
+
"loss": 2.0847,
|
1562 |
+
"step": 2260
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 0.08,
|
1566 |
+
"learning_rate": 0.00019092000000000002,
|
1567 |
+
"loss": 2.1092,
|
1568 |
+
"step": 2270
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 0.08,
|
1572 |
+
"learning_rate": 0.00019088000000000002,
|
1573 |
+
"loss": 2.1634,
|
1574 |
+
"step": 2280
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 0.08,
|
1578 |
+
"learning_rate": 0.00019084,
|
1579 |
+
"loss": 2.048,
|
1580 |
+
"step": 2290
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 0.08,
|
1584 |
+
"learning_rate": 0.0001908,
|
1585 |
+
"loss": 2.0382,
|
1586 |
+
"step": 2300
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 0.08,
|
1590 |
+
"eval_accuracy": 0.6545202576016956,
|
1591 |
+
"eval_loss": 1.7376079559326172,
|
1592 |
+
"eval_runtime": 46.305,
|
1593 |
+
"eval_samples_per_second": 9.135,
|
1594 |
+
"eval_steps_per_second": 1.145,
|
1595 |
+
"step": 2300
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 0.09,
|
1599 |
+
"learning_rate": 0.00019076,
|
1600 |
+
"loss": 2.0105,
|
1601 |
+
"step": 2310
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 0.09,
|
1605 |
+
"learning_rate": 0.00019072000000000002,
|
1606 |
+
"loss": 1.9905,
|
1607 |
+
"step": 2320
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 0.09,
|
1611 |
+
"learning_rate": 0.00019068,
|
1612 |
+
"loss": 2.1397,
|
1613 |
+
"step": 2330
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 0.09,
|
1617 |
+
"learning_rate": 0.00019064000000000003,
|
1618 |
+
"loss": 1.9861,
|
1619 |
+
"step": 2340
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.09,
|
1623 |
+
"learning_rate": 0.0001906,
|
1624 |
+
"loss": 2.0603,
|
1625 |
+
"step": 2350
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 0.09,
|
1629 |
+
"learning_rate": 0.00019056000000000002,
|
1630 |
+
"loss": 2.0137,
|
1631 |
+
"step": 2360
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 0.09,
|
1635 |
+
"learning_rate": 0.00019052,
|
1636 |
+
"loss": 2.0223,
|
1637 |
+
"step": 2370
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 0.09,
|
1641 |
+
"learning_rate": 0.00019048,
|
1642 |
+
"loss": 2.0917,
|
1643 |
+
"step": 2380
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 0.09,
|
1647 |
+
"learning_rate": 0.00019044000000000003,
|
1648 |
+
"loss": 2.0426,
|
1649 |
+
"step": 2390
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 0.09,
|
1653 |
+
"learning_rate": 0.0001904,
|
1654 |
+
"loss": 2.0427,
|
1655 |
+
"step": 2400
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 0.09,
|
1659 |
+
"eval_accuracy": 0.6534605037906579,
|
1660 |
+
"eval_loss": 1.7286646366119385,
|
1661 |
+
"eval_runtime": 46.0656,
|
1662 |
+
"eval_samples_per_second": 9.183,
|
1663 |
+
"eval_steps_per_second": 1.151,
|
1664 |
+
"step": 2400
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 0.09,
|
1668 |
+
"learning_rate": 0.00019036000000000001,
|
1669 |
+
"loss": 1.9983,
|
1670 |
+
"step": 2410
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 0.09,
|
1674 |
+
"learning_rate": 0.00019032,
|
1675 |
+
"loss": 2.0404,
|
1676 |
+
"step": 2420
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 0.09,
|
1680 |
+
"learning_rate": 0.00019028,
|
1681 |
+
"loss": 2.0165,
|
1682 |
+
"step": 2430
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.09,
|
1686 |
+
"learning_rate": 0.00019024000000000002,
|
1687 |
+
"loss": 2.0585,
|
1688 |
+
"step": 2440
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 0.09,
|
1692 |
+
"learning_rate": 0.0001902,
|
1693 |
+
"loss": 2.0517,
|
1694 |
+
"step": 2450
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 0.09,
|
1698 |
+
"learning_rate": 0.00019016,
|
1699 |
+
"loss": 2.041,
|
1700 |
+
"step": 2460
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 0.09,
|
1704 |
+
"learning_rate": 0.00019012,
|
1705 |
+
"loss": 1.9899,
|
1706 |
+
"step": 2470
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 0.09,
|
1710 |
+
"learning_rate": 0.00019008000000000002,
|
1711 |
+
"loss": 1.9965,
|
1712 |
+
"step": 2480
|
1713 |
+
},
|
1714 |
+
{
|
1715 |
+
"epoch": 0.09,
|
1716 |
+
"learning_rate": 0.00019004000000000002,
|
1717 |
+
"loss": 2.0312,
|
1718 |
+
"step": 2490
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 0.09,
|
1722 |
+
"learning_rate": 0.00019,
|
1723 |
+
"loss": 1.9543,
|
1724 |
+
"step": 2500
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.09,
|
1728 |
+
"eval_accuracy": 0.6536235428385099,
|
1729 |
+
"eval_loss": 1.7141377925872803,
|
1730 |
+
"eval_runtime": 46.1754,
|
1731 |
+
"eval_samples_per_second": 9.161,
|
1732 |
+
"eval_steps_per_second": 1.148,
|
1733 |
+
"step": 2500
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 0.09,
|
1737 |
+
"learning_rate": 0.00018996,
|
1738 |
+
"loss": 2.0157,
|
1739 |
+
"step": 2510
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 0.09,
|
1743 |
+
"learning_rate": 0.00018992,
|
1744 |
+
"loss": 2.0398,
|
1745 |
+
"step": 2520
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.09,
|
1749 |
+
"learning_rate": 0.00018988000000000002,
|
1750 |
+
"loss": 2.0151,
|
1751 |
+
"step": 2530
|
1752 |
+
},
|
1753 |
+
{
|
1754 |
+
"epoch": 0.09,
|
1755 |
+
"learning_rate": 0.00018984,
|
1756 |
+
"loss": 2.0104,
|
1757 |
+
"step": 2540
|
1758 |
+
},
|
1759 |
+
{
|
1760 |
+
"epoch": 0.09,
|
1761 |
+
"learning_rate": 0.0001898,
|
1762 |
+
"loss": 2.1258,
|
1763 |
+
"step": 2550
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 0.09,
|
1767 |
+
"learning_rate": 0.00018976,
|
1768 |
+
"loss": 2.0673,
|
1769 |
+
"step": 2560
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 0.09,
|
1773 |
+
"learning_rate": 0.00018972000000000002,
|
1774 |
+
"loss": 1.9949,
|
1775 |
+
"step": 2570
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 0.1,
|
1779 |
+
"learning_rate": 0.00018968,
|
1780 |
+
"loss": 2.0162,
|
1781 |
+
"step": 2580
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 0.1,
|
1785 |
+
"learning_rate": 0.00018964,
|
1786 |
+
"loss": 2.065,
|
1787 |
+
"step": 2590
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.1,
|
1791 |
+
"learning_rate": 0.0001896,
|
1792 |
+
"loss": 1.9798,
|
1793 |
+
"step": 2600
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 0.1,
|
1797 |
+
"eval_accuracy": 0.6568843237955491,
|
1798 |
+
"eval_loss": 1.7275162935256958,
|
1799 |
+
"eval_runtime": 46.0053,
|
1800 |
+
"eval_samples_per_second": 9.195,
|
1801 |
+
"eval_steps_per_second": 1.152,
|
1802 |
+
"step": 2600
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 0.1,
|
1806 |
+
"learning_rate": 0.00018956,
|
1807 |
+
"loss": 2.0334,
|
1808 |
+
"step": 2610
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.1,
|
1812 |
+
"learning_rate": 0.00018952000000000002,
|
1813 |
+
"loss": 2.0242,
|
1814 |
+
"step": 2620
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 0.1,
|
1818 |
+
"learning_rate": 0.00018948,
|
1819 |
+
"loss": 2.0321,
|
1820 |
+
"step": 2630
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 0.1,
|
1824 |
+
"learning_rate": 0.00018944000000000003,
|
1825 |
+
"loss": 2.0455,
|
1826 |
+
"step": 2640
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 0.1,
|
1830 |
+
"learning_rate": 0.0001894,
|
1831 |
+
"loss": 2.0152,
|
1832 |
+
"step": 2650
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 0.1,
|
1836 |
+
"learning_rate": 0.00018936000000000002,
|
1837 |
+
"loss": 2.0764,
|
1838 |
+
"step": 2660
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 0.1,
|
1842 |
+
"learning_rate": 0.00018932,
|
1843 |
+
"loss": 2.0572,
|
1844 |
+
"step": 2670
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 0.1,
|
1848 |
+
"learning_rate": 0.00018928,
|
1849 |
+
"loss": 1.9985,
|
1850 |
+
"step": 2680
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.1,
|
1854 |
+
"learning_rate": 0.00018924000000000002,
|
1855 |
+
"loss": 2.0185,
|
1856 |
+
"step": 2690
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 0.1,
|
1860 |
+
"learning_rate": 0.0001892,
|
1861 |
+
"loss": 2.0054,
|
1862 |
+
"step": 2700
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 0.1,
|
1866 |
+
"eval_accuracy": 0.6598190266568843,
|
1867 |
+
"eval_loss": 1.6908519268035889,
|
1868 |
+
"eval_runtime": 45.7104,
|
1869 |
+
"eval_samples_per_second": 9.254,
|
1870 |
+
"eval_steps_per_second": 1.159,
|
1871 |
+
"step": 2700
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.1,
|
1875 |
+
"learning_rate": 0.00018916,
|
1876 |
+
"loss": 2.0133,
|
1877 |
+
"step": 2710
|
1878 |
+
},
|
1879 |
+
{
|
1880 |
+
"epoch": 0.1,
|
1881 |
+
"learning_rate": 0.00018912,
|
1882 |
+
"loss": 2.0096,
|
1883 |
+
"step": 2720
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 0.1,
|
1887 |
+
"learning_rate": 0.00018908000000000003,
|
1888 |
+
"loss": 2.0126,
|
1889 |
+
"step": 2730
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 0.1,
|
1893 |
+
"learning_rate": 0.00018904000000000002,
|
1894 |
+
"loss": 1.9772,
|
1895 |
+
"step": 2740
|
1896 |
+
},
|
1897 |
+
{
|
1898 |
+
"epoch": 0.1,
|
1899 |
+
"learning_rate": 0.00018899999999999999,
|
1900 |
+
"loss": 2.0467,
|
1901 |
+
"step": 2750
|
1902 |
+
},
|
1903 |
+
{
|
1904 |
+
"epoch": 0.1,
|
1905 |
+
"learning_rate": 0.00018896,
|
1906 |
+
"loss": 1.9178,
|
1907 |
+
"step": 2760
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 0.1,
|
1911 |
+
"learning_rate": 0.00018892,
|
1912 |
+
"loss": 2.0297,
|
1913 |
+
"step": 2770
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.1,
|
1917 |
+
"learning_rate": 0.00018888000000000002,
|
1918 |
+
"loss": 2.0253,
|
1919 |
+
"step": 2780
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 0.1,
|
1923 |
+
"learning_rate": 0.00018884000000000001,
|
1924 |
+
"loss": 1.9633,
|
1925 |
+
"step": 2790
|
1926 |
+
},
|
1927 |
+
{
|
1928 |
+
"epoch": 0.1,
|
1929 |
+
"learning_rate": 0.0001888,
|
1930 |
+
"loss": 2.0391,
|
1931 |
+
"step": 2800
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 0.1,
|
1935 |
+
"eval_accuracy": 0.657128882367327,
|
1936 |
+
"eval_loss": 1.7396591901779175,
|
1937 |
+
"eval_runtime": 46.5669,
|
1938 |
+
"eval_samples_per_second": 9.084,
|
1939 |
+
"eval_steps_per_second": 1.138,
|
1940 |
+
"step": 2800
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 0.1,
|
1944 |
+
"learning_rate": 0.00018876,
|
1945 |
+
"loss": 2.0096,
|
1946 |
+
"step": 2810
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 0.1,
|
1950 |
+
"learning_rate": 0.00018872,
|
1951 |
+
"loss": 1.9861,
|
1952 |
+
"step": 2820
|
1953 |
+
},
|
1954 |
+
{
|
1955 |
+
"epoch": 0.1,
|
1956 |
+
"learning_rate": 0.00018868000000000002,
|
1957 |
+
"loss": 1.9337,
|
1958 |
+
"step": 2830
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 0.1,
|
1962 |
+
"learning_rate": 0.00018864,
|
1963 |
+
"loss": 2.0316,
|
1964 |
+
"step": 2840
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 0.11,
|
1968 |
+
"learning_rate": 0.0001886,
|
1969 |
+
"loss": 2.0544,
|
1970 |
+
"step": 2850
|
1971 |
+
},
|
1972 |
+
{
|
1973 |
+
"epoch": 0.11,
|
1974 |
+
"learning_rate": 0.00018856,
|
1975 |
+
"loss": 1.9592,
|
1976 |
+
"step": 2860
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.11,
|
1980 |
+
"learning_rate": 0.00018852000000000002,
|
1981 |
+
"loss": 1.9677,
|
1982 |
+
"step": 2870
|
1983 |
+
},
|
1984 |
+
{
|
1985 |
+
"epoch": 0.11,
|
1986 |
+
"learning_rate": 0.00018848,
|
1987 |
+
"loss": 2.0595,
|
1988 |
+
"step": 2880
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 0.11,
|
1992 |
+
"learning_rate": 0.00018844,
|
1993 |
+
"loss": 1.9924,
|
1994 |
+
"step": 2890
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 0.11,
|
1998 |
+
"learning_rate": 0.0001884,
|
1999 |
+
"loss": 1.988,
|
2000 |
+
"step": 2900
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 0.11,
|
2004 |
+
"eval_accuracy": 0.6597375071329583,
|
2005 |
+
"eval_loss": 1.7008874416351318,
|
2006 |
+
"eval_runtime": 46.6514,
|
2007 |
+
"eval_samples_per_second": 9.067,
|
2008 |
+
"eval_steps_per_second": 1.136,
|
2009 |
+
"step": 2900
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 0.11,
|
2013 |
+
"learning_rate": 0.00018836,
|
2014 |
+
"loss": 2.0202,
|
2015 |
+
"step": 2910
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 0.11,
|
2019 |
+
"learning_rate": 0.00018832,
|
2020 |
+
"loss": 1.9864,
|
2021 |
+
"step": 2920
|
2022 |
+
},
|
2023 |
+
{
|
2024 |
+
"epoch": 0.11,
|
2025 |
+
"learning_rate": 0.00018828,
|
2026 |
+
"loss": 1.9831,
|
2027 |
+
"step": 2930
|
2028 |
+
},
|
2029 |
+
{
|
2030 |
+
"epoch": 0.11,
|
2031 |
+
"learning_rate": 0.00018824000000000003,
|
2032 |
+
"loss": 1.9967,
|
2033 |
+
"step": 2940
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 0.11,
|
2037 |
+
"learning_rate": 0.0001882,
|
2038 |
+
"loss": 1.9784,
|
2039 |
+
"step": 2950
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.11,
|
2043 |
+
"learning_rate": 0.00018816000000000001,
|
2044 |
+
"loss": 1.9662,
|
2045 |
+
"step": 2960
|
2046 |
+
},
|
2047 |
+
{
|
2048 |
+
"epoch": 0.11,
|
2049 |
+
"learning_rate": 0.00018812,
|
2050 |
+
"loss": 1.9302,
|
2051 |
+
"step": 2970
|
2052 |
+
},
|
2053 |
+
{
|
2054 |
+
"epoch": 0.11,
|
2055 |
+
"learning_rate": 0.00018808,
|
2056 |
+
"loss": 2.0279,
|
2057 |
+
"step": 2980
|
2058 |
+
},
|
2059 |
+
{
|
2060 |
+
"epoch": 0.11,
|
2061 |
+
"learning_rate": 0.00018804000000000002,
|
2062 |
+
"loss": 2.0217,
|
2063 |
+
"step": 2990
|
2064 |
+
},
|
2065 |
+
{
|
2066 |
+
"epoch": 0.11,
|
2067 |
+
"learning_rate": 0.000188,
|
2068 |
+
"loss": 2.0044,
|
2069 |
+
"step": 3000
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 0.11,
|
2073 |
+
"eval_accuracy": 0.6644656395206652,
|
2074 |
+
"eval_loss": 1.6991885900497437,
|
2075 |
+
"eval_runtime": 45.8483,
|
2076 |
+
"eval_samples_per_second": 9.226,
|
2077 |
+
"eval_steps_per_second": 1.156,
|
2078 |
+
"step": 3000
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 0.11,
|
2082 |
+
"learning_rate": 0.00018796,
|
2083 |
+
"loss": 2.0481,
|
2084 |
+
"step": 3010
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 0.11,
|
2088 |
+
"learning_rate": 0.00018792,
|
2089 |
+
"loss": 1.9341,
|
2090 |
+
"step": 3020
|
2091 |
+
},
|
2092 |
+
{
|
2093 |
+
"epoch": 0.11,
|
2094 |
+
"learning_rate": 0.00018788000000000002,
|
2095 |
+
"loss": 1.9922,
|
2096 |
+
"step": 3030
|
2097 |
+
},
|
2098 |
+
{
|
2099 |
+
"epoch": 0.11,
|
2100 |
+
"learning_rate": 0.00018784000000000002,
|
2101 |
+
"loss": 1.9681,
|
2102 |
+
"step": 3040
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.11,
|
2106 |
+
"learning_rate": 0.0001878,
|
2107 |
+
"loss": 1.97,
|
2108 |
+
"step": 3050
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 0.11,
|
2112 |
+
"learning_rate": 0.00018776,
|
2113 |
+
"loss": 1.9463,
|
2114 |
+
"step": 3060
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 0.11,
|
2118 |
+
"learning_rate": 0.00018772,
|
2119 |
+
"loss": 1.9919,
|
2120 |
+
"step": 3070
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 0.11,
|
2124 |
+
"learning_rate": 0.00018768000000000002,
|
2125 |
+
"loss": 1.9762,
|
2126 |
+
"step": 3080
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 0.11,
|
2130 |
+
"learning_rate": 0.00018764,
|
2131 |
+
"loss": 1.9491,
|
2132 |
+
"step": 3090
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 0.11,
|
2136 |
+
"learning_rate": 0.0001876,
|
2137 |
+
"loss": 2.0099,
|
2138 |
+
"step": 3100
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 0.11,
|
2142 |
+
"eval_accuracy": 0.6629167685660716,
|
2143 |
+
"eval_loss": 1.6818691492080688,
|
2144 |
+
"eval_runtime": 47.0155,
|
2145 |
+
"eval_samples_per_second": 8.997,
|
2146 |
+
"eval_steps_per_second": 1.127,
|
2147 |
+
"step": 3100
|
2148 |
+
},
|
2149 |
+
{
|
2150 |
+
"epoch": 0.11,
|
2151 |
+
"learning_rate": 0.00018756,
|
2152 |
+
"loss": 1.9804,
|
2153 |
+
"step": 3110
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 0.12,
|
2157 |
+
"learning_rate": 0.00018752,
|
2158 |
+
"loss": 1.9109,
|
2159 |
+
"step": 3120
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 0.12,
|
2163 |
+
"learning_rate": 0.00018748000000000001,
|
2164 |
+
"loss": 1.9892,
|
2165 |
+
"step": 3130
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.12,
|
2169 |
+
"learning_rate": 0.00018744,
|
2170 |
+
"loss": 1.9546,
|
2171 |
+
"step": 3140
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 0.12,
|
2175 |
+
"learning_rate": 0.00018740000000000003,
|
2176 |
+
"loss": 1.9803,
|
2177 |
+
"step": 3150
|
2178 |
+
},
|
2179 |
+
{
|
2180 |
+
"epoch": 0.12,
|
2181 |
+
"learning_rate": 0.00018736,
|
2182 |
+
"loss": 1.9629,
|
2183 |
+
"step": 3160
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 0.12,
|
2187 |
+
"learning_rate": 0.00018732000000000002,
|
2188 |
+
"loss": 1.9725,
|
2189 |
+
"step": 3170
|
2190 |
+
},
|
2191 |
+
{
|
2192 |
+
"epoch": 0.12,
|
2193 |
+
"learning_rate": 0.00018728,
|
2194 |
+
"loss": 2.0243,
|
2195 |
+
"step": 3180
|
2196 |
+
},
|
2197 |
+
{
|
2198 |
+
"epoch": 0.12,
|
2199 |
+
"learning_rate": 0.00018724,
|
2200 |
+
"loss": 2.0426,
|
2201 |
+
"step": 3190
|
2202 |
+
},
|
2203 |
+
{
|
2204 |
+
"epoch": 0.12,
|
2205 |
+
"learning_rate": 0.00018720000000000002,
|
2206 |
+
"loss": 1.9622,
|
2207 |
+
"step": 3200
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 0.12,
|
2211 |
+
"eval_accuracy": 0.6634058857096274,
|
2212 |
+
"eval_loss": 1.6796367168426514,
|
2213 |
+
"eval_runtime": 45.6554,
|
2214 |
+
"eval_samples_per_second": 9.265,
|
2215 |
+
"eval_steps_per_second": 1.161,
|
2216 |
+
"step": 3200
|
2217 |
+
},
|
2218 |
+
{
|
2219 |
+
"epoch": 0.12,
|
2220 |
+
"learning_rate": 0.00018716,
|
2221 |
+
"loss": 2.0019,
|
2222 |
+
"step": 3210
|
2223 |
+
},
|
2224 |
+
{
|
2225 |
+
"epoch": 0.12,
|
2226 |
+
"learning_rate": 0.00018712,
|
2227 |
+
"loss": 1.9587,
|
2228 |
+
"step": 3220
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 0.12,
|
2232 |
+
"learning_rate": 0.00018708,
|
2233 |
+
"loss": 1.9816,
|
2234 |
+
"step": 3230
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 0.12,
|
2238 |
+
"learning_rate": 0.00018704000000000003,
|
2239 |
+
"loss": 1.9877,
|
2240 |
+
"step": 3240
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 0.12,
|
2244 |
+
"learning_rate": 0.00018700000000000002,
|
2245 |
+
"loss": 2.0077,
|
2246 |
+
"step": 3250
|
2247 |
+
},
|
2248 |
+
{
|
2249 |
+
"epoch": 0.12,
|
2250 |
+
"learning_rate": 0.00018696,
|
2251 |
+
"loss": 2.043,
|
2252 |
+
"step": 3260
|
2253 |
+
},
|
2254 |
+
{
|
2255 |
+
"epoch": 0.12,
|
2256 |
+
"learning_rate": 0.00018692,
|
2257 |
+
"loss": 1.9861,
|
2258 |
+
"step": 3270
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 0.12,
|
2262 |
+
"learning_rate": 0.00018688,
|
2263 |
+
"loss": 1.9399,
|
2264 |
+
"step": 3280
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 0.12,
|
2268 |
+
"learning_rate": 0.00018684000000000002,
|
2269 |
+
"loss": 1.9473,
|
2270 |
+
"step": 3290
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 0.12,
|
2274 |
+
"learning_rate": 0.00018680000000000001,
|
2275 |
+
"loss": 1.9716,
|
2276 |
+
"step": 3300
|
2277 |
+
},
|
2278 |
+
{
|
2279 |
+
"epoch": 0.12,
|
2280 |
+
"eval_accuracy": 0.657128882367327,
|
2281 |
+
"eval_loss": 1.7205617427825928,
|
2282 |
+
"eval_runtime": 46.0038,
|
2283 |
+
"eval_samples_per_second": 9.195,
|
2284 |
+
"eval_steps_per_second": 1.152,
|
2285 |
+
"step": 3300
|
2286 |
+
},
|
2287 |
+
{
|
2288 |
+
"epoch": 0.12,
|
2289 |
+
"learning_rate": 0.00018676,
|
2290 |
+
"loss": 1.9388,
|
2291 |
+
"step": 3310
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 0.12,
|
2295 |
+
"learning_rate": 0.00018672,
|
2296 |
+
"loss": 1.9062,
|
2297 |
+
"step": 3320
|
2298 |
+
},
|
2299 |
+
{
|
2300 |
+
"epoch": 0.12,
|
2301 |
+
"learning_rate": 0.00018668000000000002,
|
2302 |
+
"loss": 2.0283,
|
2303 |
+
"step": 3330
|
2304 |
+
},
|
2305 |
+
{
|
2306 |
+
"epoch": 0.12,
|
2307 |
+
"learning_rate": 0.00018664000000000002,
|
2308 |
+
"loss": 1.9347,
|
2309 |
+
"step": 3340
|
2310 |
+
},
|
2311 |
+
{
|
2312 |
+
"epoch": 0.12,
|
2313 |
+
"learning_rate": 0.0001866,
|
2314 |
+
"loss": 1.9422,
|
2315 |
+
"step": 3350
|
2316 |
+
},
|
2317 |
+
{
|
2318 |
+
"epoch": 0.12,
|
2319 |
+
"learning_rate": 0.00018656,
|
2320 |
+
"loss": 1.935,
|
2321 |
+
"step": 3360
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 0.12,
|
2325 |
+
"learning_rate": 0.00018652,
|
2326 |
+
"loss": 1.9642,
|
2327 |
+
"step": 3370
|
2328 |
+
},
|
2329 |
+
{
|
2330 |
+
"epoch": 0.12,
|
2331 |
+
"learning_rate": 0.00018648000000000002,
|
2332 |
+
"loss": 1.9179,
|
2333 |
+
"step": 3380
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 0.13,
|
2337 |
+
"learning_rate": 0.00018644,
|
2338 |
+
"loss": 1.9948,
|
2339 |
+
"step": 3390
|
2340 |
+
},
|
2341 |
+
{
|
2342 |
+
"epoch": 0.13,
|
2343 |
+
"learning_rate": 0.00018640000000000003,
|
2344 |
+
"loss": 1.9136,
|
2345 |
+
"step": 3400
|
2346 |
+
},
|
2347 |
+
{
|
2348 |
+
"epoch": 0.13,
|
2349 |
+
"eval_accuracy": 0.6654438738077769,
|
2350 |
+
"eval_loss": 1.6819649934768677,
|
2351 |
+
"eval_runtime": 45.6304,
|
2352 |
+
"eval_samples_per_second": 9.27,
|
2353 |
+
"eval_steps_per_second": 1.162,
|
2354 |
+
"step": 3400
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 0.13,
|
2358 |
+
"learning_rate": 0.00018636,
|
2359 |
+
"loss": 1.8923,
|
2360 |
+
"step": 3410
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 0.13,
|
2364 |
+
"learning_rate": 0.00018632000000000002,
|
2365 |
+
"loss": 1.9587,
|
2366 |
+
"step": 3420
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 0.13,
|
2370 |
+
"learning_rate": 0.00018628,
|
2371 |
+
"loss": 1.9211,
|
2372 |
+
"step": 3430
|
2373 |
+
},
|
2374 |
+
{
|
2375 |
+
"epoch": 0.13,
|
2376 |
+
"learning_rate": 0.00018624,
|
2377 |
+
"loss": 1.9302,
|
2378 |
+
"step": 3440
|
2379 |
+
},
|
2380 |
+
{
|
2381 |
+
"epoch": 0.13,
|
2382 |
+
"learning_rate": 0.00018620000000000003,
|
2383 |
+
"loss": 1.9764,
|
2384 |
+
"step": 3450
|
2385 |
+
},
|
2386 |
+
{
|
2387 |
+
"epoch": 0.13,
|
2388 |
+
"learning_rate": 0.00018616,
|
2389 |
+
"loss": 1.9956,
|
2390 |
+
"step": 3460
|
2391 |
+
},
|
2392 |
+
{
|
2393 |
+
"epoch": 0.13,
|
2394 |
+
"learning_rate": 0.00018612000000000001,
|
2395 |
+
"loss": 1.9196,
|
2396 |
+
"step": 3470
|
2397 |
+
},
|
2398 |
+
{
|
2399 |
+
"epoch": 0.13,
|
2400 |
+
"learning_rate": 0.00018608,
|
2401 |
+
"loss": 1.955,
|
2402 |
+
"step": 3480
|
2403 |
+
},
|
2404 |
+
{
|
2405 |
+
"epoch": 0.13,
|
2406 |
+
"learning_rate": 0.00018604,
|
2407 |
+
"loss": 1.9914,
|
2408 |
+
"step": 3490
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 0.13,
|
2412 |
+
"learning_rate": 0.00018600000000000002,
|
2413 |
+
"loss": 1.9277,
|
2414 |
+
"step": 3500
|
2415 |
+
},
|
2416 |
+
{
|
2417 |
+
"epoch": 0.13,
|
2418 |
+
"eval_accuracy": 0.6650362761881471,
|
2419 |
+
"eval_loss": 1.6978471279144287,
|
2420 |
+
"eval_runtime": 45.6548,
|
2421 |
+
"eval_samples_per_second": 9.265,
|
2422 |
+
"eval_steps_per_second": 1.161,
|
2423 |
+
"step": 3500
|
2424 |
+
},
|
2425 |
+
{
|
2426 |
+
"epoch": 0.13,
|
2427 |
+
"learning_rate": 0.00018596,
|
2428 |
+
"loss": 1.949,
|
2429 |
+
"step": 3510
|
2430 |
+
},
|
2431 |
+
{
|
2432 |
+
"epoch": 0.13,
|
2433 |
+
"learning_rate": 0.00018592,
|
2434 |
+
"loss": 1.927,
|
2435 |
+
"step": 3520
|
2436 |
+
},
|
2437 |
+
{
|
2438 |
+
"epoch": 0.13,
|
2439 |
+
"learning_rate": 0.00018588,
|
2440 |
+
"loss": 2.0058,
|
2441 |
+
"step": 3530
|
2442 |
+
},
|
2443 |
+
{
|
2444 |
+
"epoch": 0.13,
|
2445 |
+
"learning_rate": 0.00018584000000000002,
|
2446 |
+
"loss": 1.9854,
|
2447 |
+
"step": 3540
|
2448 |
+
},
|
2449 |
+
{
|
2450 |
+
"epoch": 0.13,
|
2451 |
+
"learning_rate": 0.00018580000000000002,
|
2452 |
+
"loss": 1.9646,
|
2453 |
+
"step": 3550
|
2454 |
+
},
|
2455 |
+
{
|
2456 |
+
"epoch": 0.13,
|
2457 |
+
"learning_rate": 0.00018576,
|
2458 |
+
"loss": 1.9517,
|
2459 |
+
"step": 3560
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 0.13,
|
2463 |
+
"learning_rate": 0.00018572,
|
2464 |
+
"loss": 2.0103,
|
2465 |
+
"step": 3570
|
2466 |
+
},
|
2467 |
+
{
|
2468 |
+
"epoch": 0.13,
|
2469 |
+
"learning_rate": 0.00018568,
|
2470 |
+
"loss": 1.9367,
|
2471 |
+
"step": 3580
|
2472 |
+
},
|
2473 |
+
{
|
2474 |
+
"epoch": 0.13,
|
2475 |
+
"learning_rate": 0.00018564000000000002,
|
2476 |
+
"loss": 1.9968,
|
2477 |
+
"step": 3590
|
2478 |
+
},
|
2479 |
+
{
|
2480 |
+
"epoch": 0.13,
|
2481 |
+
"learning_rate": 0.0001856,
|
2482 |
+
"loss": 1.9727,
|
2483 |
+
"step": 3600
|
2484 |
+
},
|
2485 |
+
{
|
2486 |
+
"epoch": 0.13,
|
2487 |
+
"eval_accuracy": 0.6593299095133285,
|
2488 |
+
"eval_loss": 1.6777493953704834,
|
2489 |
+
"eval_runtime": 60.7874,
|
2490 |
+
"eval_samples_per_second": 6.959,
|
2491 |
+
"eval_steps_per_second": 0.872,
|
2492 |
+
"step": 3600
|
2493 |
+
},
|
2494 |
+
{
|
2495 |
+
"epoch": 0.13,
|
2496 |
+
"learning_rate": 0.00018556,
|
2497 |
+
"loss": 1.9233,
|
2498 |
+
"step": 3610
|
2499 |
+
},
|
2500 |
+
{
|
2501 |
+
"epoch": 0.13,
|
2502 |
+
"learning_rate": 0.00018552,
|
2503 |
+
"loss": 1.926,
|
2504 |
+
"step": 3620
|
2505 |
+
},
|
2506 |
+
{
|
2507 |
+
"epoch": 0.13,
|
2508 |
+
"learning_rate": 0.00018548000000000002,
|
2509 |
+
"loss": 1.9538,
|
2510 |
+
"step": 3630
|
2511 |
+
},
|
2512 |
+
{
|
2513 |
+
"epoch": 0.13,
|
2514 |
+
"learning_rate": 0.00018544,
|
2515 |
+
"loss": 1.9795,
|
2516 |
+
"step": 3640
|
2517 |
+
},
|
2518 |
+
{
|
2519 |
+
"epoch": 0.13,
|
2520 |
+
"learning_rate": 0.0001854,
|
2521 |
+
"loss": 1.9675,
|
2522 |
+
"step": 3650
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 0.14,
|
2526 |
+
"learning_rate": 0.00018536,
|
2527 |
+
"loss": 1.8831,
|
2528 |
+
"step": 3660
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 0.14,
|
2532 |
+
"learning_rate": 0.00018532,
|
2533 |
+
"loss": 1.9082,
|
2534 |
+
"step": 3670
|
2535 |
+
},
|
2536 |
+
{
|
2537 |
+
"epoch": 0.14,
|
2538 |
+
"learning_rate": 0.00018528000000000001,
|
2539 |
+
"loss": 1.9514,
|
2540 |
+
"step": 3680
|
2541 |
+
},
|
2542 |
+
{
|
2543 |
+
"epoch": 0.14,
|
2544 |
+
"learning_rate": 0.00018524,
|
2545 |
+
"loss": 1.9714,
|
2546 |
+
"step": 3690
|
2547 |
+
},
|
2548 |
+
{
|
2549 |
+
"epoch": 0.14,
|
2550 |
+
"learning_rate": 0.00018520000000000003,
|
2551 |
+
"loss": 2.0391,
|
2552 |
+
"step": 3700
|
2553 |
+
},
|
2554 |
+
{
|
2555 |
+
"epoch": 0.14,
|
2556 |
+
"eval_accuracy": 0.6623461318985897,
|
2557 |
+
"eval_loss": 1.6935285329818726,
|
2558 |
+
"eval_runtime": 59.3856,
|
2559 |
+
"eval_samples_per_second": 7.123,
|
2560 |
+
"eval_steps_per_second": 0.892,
|
2561 |
+
"step": 3700
|
2562 |
+
},
|
2563 |
+
{
|
2564 |
+
"epoch": 0.14,
|
2565 |
+
"learning_rate": 0.00018516,
|
2566 |
+
"loss": 1.9065,
|
2567 |
+
"step": 3710
|
2568 |
+
},
|
2569 |
+
{
|
2570 |
+
"epoch": 0.14,
|
2571 |
+
"learning_rate": 0.00018512000000000002,
|
2572 |
+
"loss": 1.9062,
|
2573 |
+
"step": 3720
|
2574 |
+
},
|
2575 |
+
{
|
2576 |
+
"epoch": 0.14,
|
2577 |
+
"learning_rate": 0.00018508,
|
2578 |
+
"loss": 1.9187,
|
2579 |
+
"step": 3730
|
2580 |
+
},
|
2581 |
+
{
|
2582 |
+
"epoch": 0.14,
|
2583 |
+
"learning_rate": 0.00018504,
|
2584 |
+
"loss": 1.9407,
|
2585 |
+
"step": 3740
|
2586 |
+
},
|
2587 |
+
{
|
2588 |
+
"epoch": 0.14,
|
2589 |
+
"learning_rate": 0.00018500000000000002,
|
2590 |
+
"loss": 1.9684,
|
2591 |
+
"step": 3750
|
2592 |
+
},
|
2593 |
+
{
|
2594 |
+
"epoch": 0.14,
|
2595 |
+
"learning_rate": 0.00018496,
|
2596 |
+
"loss": 1.9124,
|
2597 |
+
"step": 3760
|
2598 |
+
},
|
2599 |
+
{
|
2600 |
+
"epoch": 0.14,
|
2601 |
+
"learning_rate": 0.00018492,
|
2602 |
+
"loss": 1.9626,
|
2603 |
+
"step": 3770
|
2604 |
+
},
|
2605 |
+
{
|
2606 |
+
"epoch": 0.14,
|
2607 |
+
"learning_rate": 0.00018488,
|
2608 |
+
"loss": 1.8665,
|
2609 |
+
"step": 3780
|
2610 |
+
},
|
2611 |
+
{
|
2612 |
+
"epoch": 0.14,
|
2613 |
+
"learning_rate": 0.00018484000000000003,
|
2614 |
+
"loss": 1.9432,
|
2615 |
+
"step": 3790
|
2616 |
+
},
|
2617 |
+
{
|
2618 |
+
"epoch": 0.14,
|
2619 |
+
"learning_rate": 0.00018480000000000002,
|
2620 |
+
"loss": 1.9367,
|
2621 |
+
"step": 3800
|
2622 |
+
},
|
2623 |
+
{
|
2624 |
+
"epoch": 0.14,
|
2625 |
+
"eval_accuracy": 0.6560691285562892,
|
2626 |
+
"eval_loss": 1.7309006452560425,
|
2627 |
+
"eval_runtime": 61.2631,
|
2628 |
+
"eval_samples_per_second": 6.905,
|
2629 |
+
"eval_steps_per_second": 0.865,
|
2630 |
+
"step": 3800
|
2631 |
+
},
|
2632 |
+
{
|
2633 |
+
"epoch": 0.14,
|
2634 |
+
"learning_rate": 0.00018476,
|
2635 |
+
"loss": 1.9483,
|
2636 |
+
"step": 3810
|
2637 |
+
},
|
2638 |
+
{
|
2639 |
+
"epoch": 0.14,
|
2640 |
+
"learning_rate": 0.00018472,
|
2641 |
+
"loss": 1.9067,
|
2642 |
+
"step": 3820
|
2643 |
+
},
|
2644 |
+
{
|
2645 |
+
"epoch": 0.14,
|
2646 |
+
"learning_rate": 0.00018468,
|
2647 |
+
"loss": 1.9485,
|
2648 |
+
"step": 3830
|
2649 |
+
},
|
2650 |
+
{
|
2651 |
+
"epoch": 0.14,
|
2652 |
+
"learning_rate": 0.00018464000000000002,
|
2653 |
+
"loss": 1.9529,
|
2654 |
+
"step": 3840
|
2655 |
+
},
|
2656 |
+
{
|
2657 |
+
"epoch": 0.14,
|
2658 |
+
"learning_rate": 0.00018460000000000001,
|
2659 |
+
"loss": 1.9463,
|
2660 |
+
"step": 3850
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 0.14,
|
2664 |
+
"learning_rate": 0.00018456,
|
2665 |
+
"loss": 2.0001,
|
2666 |
+
"step": 3860
|
2667 |
+
},
|
2668 |
+
{
|
2669 |
+
"epoch": 0.14,
|
2670 |
+
"learning_rate": 0.00018452,
|
2671 |
+
"loss": 1.8994,
|
2672 |
+
"step": 3870
|
2673 |
+
},
|
2674 |
+
{
|
2675 |
+
"epoch": 0.14,
|
2676 |
+
"learning_rate": 0.00018448,
|
2677 |
+
"loss": 1.8648,
|
2678 |
+
"step": 3880
|
2679 |
+
},
|
2680 |
+
{
|
2681 |
+
"epoch": 0.14,
|
2682 |
+
"learning_rate": 0.00018444000000000002,
|
2683 |
+
"loss": 1.9028,
|
2684 |
+
"step": 3890
|
2685 |
+
},
|
2686 |
+
{
|
2687 |
+
"epoch": 0.14,
|
2688 |
+
"learning_rate": 0.0001844,
|
2689 |
+
"loss": 1.9146,
|
2690 |
+
"step": 3900
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 0.14,
|
2694 |
+
"eval_accuracy": 0.6616939757071819,
|
2695 |
+
"eval_loss": 1.7054038047790527,
|
2696 |
+
"eval_runtime": 60.7176,
|
2697 |
+
"eval_samples_per_second": 6.967,
|
2698 |
+
"eval_steps_per_second": 0.873,
|
2699 |
+
"step": 3900
|
2700 |
+
},
|
2701 |
+
{
|
2702 |
+
"epoch": 0.14,
|
2703 |
+
"learning_rate": 0.00018436,
|
2704 |
+
"loss": 1.945,
|
2705 |
+
"step": 3910
|
2706 |
+
},
|
2707 |
+
{
|
2708 |
+
"epoch": 0.14,
|
2709 |
+
"learning_rate": 0.00018432,
|
2710 |
+
"loss": 1.9027,
|
2711 |
+
"step": 3920
|
2712 |
+
},
|
2713 |
+
{
|
2714 |
+
"epoch": 0.15,
|
2715 |
+
"learning_rate": 0.00018428000000000002,
|
2716 |
+
"loss": 1.959,
|
2717 |
+
"step": 3930
|
2718 |
+
},
|
2719 |
+
{
|
2720 |
+
"epoch": 0.15,
|
2721 |
+
"learning_rate": 0.00018424,
|
2722 |
+
"loss": 1.9469,
|
2723 |
+
"step": 3940
|
2724 |
+
},
|
2725 |
+
{
|
2726 |
+
"epoch": 0.15,
|
2727 |
+
"learning_rate": 0.0001842,
|
2728 |
+
"loss": 1.9149,
|
2729 |
+
"step": 3950
|
2730 |
+
},
|
2731 |
+
{
|
2732 |
+
"epoch": 0.15,
|
2733 |
+
"learning_rate": 0.00018416,
|
2734 |
+
"loss": 1.9151,
|
2735 |
+
"step": 3960
|
2736 |
+
},
|
2737 |
+
{
|
2738 |
+
"epoch": 0.15,
|
2739 |
+
"learning_rate": 0.00018412,
|
2740 |
+
"loss": 1.9054,
|
2741 |
+
"step": 3970
|
2742 |
+
},
|
2743 |
+
{
|
2744 |
+
"epoch": 0.15,
|
2745 |
+
"learning_rate": 0.00018408,
|
2746 |
+
"loss": 1.8939,
|
2747 |
+
"step": 3980
|
2748 |
+
},
|
2749 |
+
{
|
2750 |
+
"epoch": 0.15,
|
2751 |
+
"learning_rate": 0.00018404,
|
2752 |
+
"loss": 1.9331,
|
2753 |
+
"step": 3990
|
2754 |
+
},
|
2755 |
+
{
|
2756 |
+
"epoch": 0.15,
|
2757 |
+
"learning_rate": 0.00018400000000000003,
|
2758 |
+
"loss": 1.957,
|
2759 |
+
"step": 4000
|
2760 |
+
},
|
2761 |
+
{
|
2762 |
+
"epoch": 0.15,
|
2763 |
+
"eval_accuracy": 0.6670742642862966,
|
2764 |
+
"eval_loss": 1.6463383436203003,
|
2765 |
+
"eval_runtime": 46.0232,
|
2766 |
+
"eval_samples_per_second": 9.191,
|
2767 |
+
"eval_steps_per_second": 1.152,
|
2768 |
+
"step": 4000
|
2769 |
+
},
|
2770 |
+
{
|
2771 |
+
"epoch": 0.15,
|
2772 |
+
"learning_rate": 0.00018396,
|
2773 |
+
"loss": 1.9497,
|
2774 |
+
"step": 4010
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 0.15,
|
2778 |
+
"learning_rate": 0.00018392000000000001,
|
2779 |
+
"loss": 1.9271,
|
2780 |
+
"step": 4020
|
2781 |
+
},
|
2782 |
+
{
|
2783 |
+
"epoch": 0.15,
|
2784 |
+
"learning_rate": 0.00018388,
|
2785 |
+
"loss": 1.9087,
|
2786 |
+
"step": 4030
|
2787 |
+
},
|
2788 |
+
{
|
2789 |
+
"epoch": 0.15,
|
2790 |
+
"learning_rate": 0.00018384,
|
2791 |
+
"loss": 1.9161,
|
2792 |
+
"step": 4040
|
2793 |
+
},
|
2794 |
+
{
|
2795 |
+
"epoch": 0.15,
|
2796 |
+
"learning_rate": 0.00018380000000000002,
|
2797 |
+
"loss": 1.9423,
|
2798 |
+
"step": 4050
|
2799 |
+
},
|
2800 |
+
{
|
2801 |
+
"epoch": 0.15,
|
2802 |
+
"learning_rate": 0.00018376,
|
2803 |
+
"loss": 1.8995,
|
2804 |
+
"step": 4060
|
2805 |
+
},
|
2806 |
+
{
|
2807 |
+
"epoch": 0.15,
|
2808 |
+
"learning_rate": 0.00018372,
|
2809 |
+
"loss": 1.9421,
|
2810 |
+
"step": 4070
|
2811 |
+
},
|
2812 |
+
{
|
2813 |
+
"epoch": 0.15,
|
2814 |
+
"learning_rate": 0.00018368,
|
2815 |
+
"loss": 1.9013,
|
2816 |
+
"step": 4080
|
2817 |
+
},
|
2818 |
+
{
|
2819 |
+
"epoch": 0.15,
|
2820 |
+
"learning_rate": 0.00018364000000000002,
|
2821 |
+
"loss": 1.9092,
|
2822 |
+
"step": 4090
|
2823 |
+
},
|
2824 |
+
{
|
2825 |
+
"epoch": 0.15,
|
2826 |
+
"learning_rate": 0.00018360000000000002,
|
2827 |
+
"loss": 1.882,
|
2828 |
+
"step": 4100
|
2829 |
+
},
|
2830 |
+
{
|
2831 |
+
"epoch": 0.15,
|
2832 |
+
"eval_accuracy": 0.6641395614249613,
|
2833 |
+
"eval_loss": 1.6783875226974487,
|
2834 |
+
"eval_runtime": 47.0079,
|
2835 |
+
"eval_samples_per_second": 8.998,
|
2836 |
+
"eval_steps_per_second": 1.127,
|
2837 |
+
"step": 4100
|
2838 |
+
},
|
2839 |
+
{
|
2840 |
+
"epoch": 0.15,
|
2841 |
+
"learning_rate": 0.00018356,
|
2842 |
+
"loss": 1.9424,
|
2843 |
+
"step": 4110
|
2844 |
+
},
|
2845 |
+
{
|
2846 |
+
"epoch": 0.15,
|
2847 |
+
"learning_rate": 0.00018352,
|
2848 |
+
"loss": 1.9477,
|
2849 |
+
"step": 4120
|
2850 |
+
},
|
2851 |
+
{
|
2852 |
+
"epoch": 0.15,
|
2853 |
+
"learning_rate": 0.00018348,
|
2854 |
+
"loss": 1.9255,
|
2855 |
+
"step": 4130
|
2856 |
+
},
|
2857 |
+
{
|
2858 |
+
"epoch": 0.15,
|
2859 |
+
"learning_rate": 0.00018344000000000002,
|
2860 |
+
"loss": 1.9456,
|
2861 |
+
"step": 4140
|
2862 |
+
},
|
2863 |
+
{
|
2864 |
+
"epoch": 0.15,
|
2865 |
+
"learning_rate": 0.0001834,
|
2866 |
+
"loss": 1.9288,
|
2867 |
+
"step": 4150
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 0.15,
|
2871 |
+
"learning_rate": 0.00018336,
|
2872 |
+
"loss": 1.9088,
|
2873 |
+
"step": 4160
|
2874 |
+
},
|
2875 |
+
{
|
2876 |
+
"epoch": 0.15,
|
2877 |
+
"learning_rate": 0.00018332,
|
2878 |
+
"loss": 1.959,
|
2879 |
+
"step": 4170
|
2880 |
+
},
|
2881 |
+
{
|
2882 |
+
"epoch": 0.15,
|
2883 |
+
"learning_rate": 0.00018328000000000002,
|
2884 |
+
"loss": 1.9619,
|
2885 |
+
"step": 4180
|
2886 |
+
},
|
2887 |
+
{
|
2888 |
+
"epoch": 0.15,
|
2889 |
+
"learning_rate": 0.00018324000000000001,
|
2890 |
+
"loss": 1.9511,
|
2891 |
+
"step": 4190
|
2892 |
+
},
|
2893 |
+
{
|
2894 |
+
"epoch": 0.16,
|
2895 |
+
"learning_rate": 0.0001832,
|
2896 |
+
"loss": 1.9489,
|
2897 |
+
"step": 4200
|
2898 |
+
},
|
2899 |
+
{
|
2900 |
+
"epoch": 0.16,
|
2901 |
+
"eval_accuracy": 0.6603081438004402,
|
2902 |
+
"eval_loss": 1.676958680152893,
|
2903 |
+
"eval_runtime": 46.3451,
|
2904 |
+
"eval_samples_per_second": 9.127,
|
2905 |
+
"eval_steps_per_second": 1.144,
|
2906 |
+
"step": 4200
|
2907 |
+
},
|
2908 |
+
{
|
2909 |
+
"epoch": 0.16,
|
2910 |
+
"learning_rate": 0.00018316,
|
2911 |
+
"loss": 1.8862,
|
2912 |
+
"step": 4210
|
2913 |
+
},
|
2914 |
+
{
|
2915 |
+
"epoch": 0.16,
|
2916 |
+
"learning_rate": 0.00018312,
|
2917 |
+
"loss": 1.8631,
|
2918 |
+
"step": 4220
|
2919 |
+
},
|
2920 |
+
{
|
2921 |
+
"epoch": 0.16,
|
2922 |
+
"learning_rate": 0.00018308000000000002,
|
2923 |
+
"loss": 1.9649,
|
2924 |
+
"step": 4230
|
2925 |
+
},
|
2926 |
+
{
|
2927 |
+
"epoch": 0.16,
|
2928 |
+
"learning_rate": 0.00018304,
|
2929 |
+
"loss": 1.9592,
|
2930 |
+
"step": 4240
|
2931 |
+
},
|
2932 |
+
{
|
2933 |
+
"epoch": 0.16,
|
2934 |
+
"learning_rate": 0.000183,
|
2935 |
+
"loss": 1.9317,
|
2936 |
+
"step": 4250
|
2937 |
+
},
|
2938 |
+
{
|
2939 |
+
"epoch": 0.16,
|
2940 |
+
"learning_rate": 0.00018296,
|
2941 |
+
"loss": 1.9138,
|
2942 |
+
"step": 4260
|
2943 |
+
},
|
2944 |
+
{
|
2945 |
+
"epoch": 0.16,
|
2946 |
+
"learning_rate": 0.00018292,
|
2947 |
+
"loss": 1.8876,
|
2948 |
+
"step": 4270
|
2949 |
+
},
|
2950 |
+
{
|
2951 |
+
"epoch": 0.16,
|
2952 |
+
"learning_rate": 0.00018288,
|
2953 |
+
"loss": 1.859,
|
2954 |
+
"step": 4280
|
2955 |
+
},
|
2956 |
+
{
|
2957 |
+
"epoch": 0.16,
|
2958 |
+
"learning_rate": 0.00018284,
|
2959 |
+
"loss": 1.9496,
|
2960 |
+
"step": 4290
|
2961 |
+
},
|
2962 |
+
{
|
2963 |
+
"epoch": 0.16,
|
2964 |
+
"learning_rate": 0.00018280000000000003,
|
2965 |
+
"loss": 1.9407,
|
2966 |
+
"step": 4300
|
2967 |
+
},
|
2968 |
+
{
|
2969 |
+
"epoch": 0.16,
|
2970 |
+
"eval_accuracy": 0.669601369528002,
|
2971 |
+
"eval_loss": 1.641465187072754,
|
2972 |
+
"eval_runtime": 46.8321,
|
2973 |
+
"eval_samples_per_second": 9.032,
|
2974 |
+
"eval_steps_per_second": 1.132,
|
2975 |
+
"step": 4300
|
2976 |
+
},
|
2977 |
+
{
|
2978 |
+
"epoch": 0.16,
|
2979 |
+
"learning_rate": 0.00018276,
|
2980 |
+
"loss": 1.9654,
|
2981 |
+
"step": 4310
|
2982 |
+
},
|
2983 |
+
{
|
2984 |
+
"epoch": 0.16,
|
2985 |
+
"learning_rate": 0.00018272,
|
2986 |
+
"loss": 1.9223,
|
2987 |
+
"step": 4320
|
2988 |
+
},
|
2989 |
+
{
|
2990 |
+
"epoch": 0.16,
|
2991 |
+
"learning_rate": 0.00018268,
|
2992 |
+
"loss": 1.9163,
|
2993 |
+
"step": 4330
|
2994 |
+
},
|
2995 |
+
{
|
2996 |
+
"epoch": 0.16,
|
2997 |
+
"learning_rate": 0.00018264,
|
2998 |
+
"loss": 1.9175,
|
2999 |
+
"step": 4340
|
3000 |
+
},
|
3001 |
+
{
|
3002 |
+
"epoch": 0.16,
|
3003 |
+
"learning_rate": 0.00018260000000000002,
|
3004 |
+
"loss": 1.9255,
|
3005 |
+
"step": 4350
|
3006 |
+
},
|
3007 |
+
{
|
3008 |
+
"epoch": 0.16,
|
3009 |
+
"learning_rate": 0.00018256,
|
3010 |
+
"loss": 1.8736,
|
3011 |
+
"step": 4360
|
3012 |
+
},
|
3013 |
+
{
|
3014 |
+
"epoch": 0.16,
|
3015 |
+
"learning_rate": 0.00018252,
|
3016 |
+
"loss": 1.8851,
|
3017 |
+
"step": 4370
|
3018 |
+
},
|
3019 |
+
{
|
3020 |
+
"epoch": 0.16,
|
3021 |
+
"learning_rate": 0.00018248,
|
3022 |
+
"loss": 1.9125,
|
3023 |
+
"step": 4380
|
3024 |
+
},
|
3025 |
+
{
|
3026 |
+
"epoch": 0.16,
|
3027 |
+
"learning_rate": 0.00018244000000000002,
|
3028 |
+
"loss": 1.888,
|
3029 |
+
"step": 4390
|
3030 |
+
},
|
3031 |
+
{
|
3032 |
+
"epoch": 0.16,
|
3033 |
+
"learning_rate": 0.00018240000000000002,
|
3034 |
+
"loss": 1.9402,
|
3035 |
+
"step": 4400
|
3036 |
+
},
|
3037 |
+
{
|
3038 |
+
"epoch": 0.16,
|
3039 |
+
"eval_accuracy": 0.6654438738077769,
|
3040 |
+
"eval_loss": 1.681833267211914,
|
3041 |
+
"eval_runtime": 46.6168,
|
3042 |
+
"eval_samples_per_second": 9.074,
|
3043 |
+
"eval_steps_per_second": 1.137,
|
3044 |
+
"step": 4400
|
3045 |
+
},
|
3046 |
+
{
|
3047 |
+
"epoch": 0.16,
|
3048 |
+
"learning_rate": 0.00018236,
|
3049 |
+
"loss": 1.8771,
|
3050 |
+
"step": 4410
|
3051 |
+
},
|
3052 |
+
{
|
3053 |
+
"epoch": 0.16,
|
3054 |
+
"learning_rate": 0.00018232,
|
3055 |
+
"loss": 1.9207,
|
3056 |
+
"step": 4420
|
3057 |
+
},
|
3058 |
+
{
|
3059 |
+
"epoch": 0.16,
|
3060 |
+
"learning_rate": 0.00018228,
|
3061 |
+
"loss": 1.8753,
|
3062 |
+
"step": 4430
|
3063 |
+
},
|
3064 |
+
{
|
3065 |
+
"epoch": 0.16,
|
3066 |
+
"learning_rate": 0.00018224000000000002,
|
3067 |
+
"loss": 1.8738,
|
3068 |
+
"step": 4440
|
3069 |
+
},
|
3070 |
+
{
|
3071 |
+
"epoch": 0.16,
|
3072 |
+
"learning_rate": 0.0001822,
|
3073 |
+
"loss": 1.9238,
|
3074 |
+
"step": 4450
|
3075 |
+
},
|
3076 |
+
{
|
3077 |
+
"epoch": 0.16,
|
3078 |
+
"learning_rate": 0.00018216000000000003,
|
3079 |
+
"loss": 1.8845,
|
3080 |
+
"step": 4460
|
3081 |
+
},
|
3082 |
+
{
|
3083 |
+
"epoch": 0.17,
|
3084 |
+
"learning_rate": 0.00018212,
|
3085 |
+
"loss": 1.9076,
|
3086 |
+
"step": 4470
|
3087 |
+
},
|
3088 |
+
{
|
3089 |
+
"epoch": 0.17,
|
3090 |
+
"learning_rate": 0.00018208000000000002,
|
3091 |
+
"loss": 1.9064,
|
3092 |
+
"step": 4480
|
3093 |
+
},
|
3094 |
+
{
|
3095 |
+
"epoch": 0.17,
|
3096 |
+
"learning_rate": 0.00018204,
|
3097 |
+
"loss": 1.9555,
|
3098 |
+
"step": 4490
|
3099 |
+
},
|
3100 |
+
{
|
3101 |
+
"epoch": 0.17,
|
3102 |
+
"learning_rate": 0.000182,
|
3103 |
+
"loss": 1.894,
|
3104 |
+
"step": 4500
|
3105 |
+
},
|
3106 |
+
{
|
3107 |
+
"epoch": 0.17,
|
3108 |
+
"eval_accuracy": 0.6659329909513328,
|
3109 |
+
"eval_loss": 1.664554476737976,
|
3110 |
+
"eval_runtime": 46.5128,
|
3111 |
+
"eval_samples_per_second": 9.094,
|
3112 |
+
"eval_steps_per_second": 1.139,
|
3113 |
+
"step": 4500
|
3114 |
+
},
|
3115 |
+
{
|
3116 |
+
"epoch": 0.17,
|
3117 |
+
"learning_rate": 0.00018196000000000003,
|
3118 |
+
"loss": 1.9158,
|
3119 |
+
"step": 4510
|
3120 |
+
},
|
3121 |
+
{
|
3122 |
+
"epoch": 0.17,
|
3123 |
+
"learning_rate": 0.00018192,
|
3124 |
+
"loss": 1.9474,
|
3125 |
+
"step": 4520
|
3126 |
+
},
|
3127 |
+
{
|
3128 |
+
"epoch": 0.17,
|
3129 |
+
"learning_rate": 0.00018188000000000001,
|
3130 |
+
"loss": 1.9463,
|
3131 |
+
"step": 4530
|
3132 |
+
},
|
3133 |
+
{
|
3134 |
+
"epoch": 0.17,
|
3135 |
+
"learning_rate": 0.00018184,
|
3136 |
+
"loss": 1.8755,
|
3137 |
+
"step": 4540
|
3138 |
+
},
|
3139 |
+
{
|
3140 |
+
"epoch": 0.17,
|
3141 |
+
"learning_rate": 0.00018180000000000003,
|
3142 |
+
"loss": 1.9637,
|
3143 |
+
"step": 4550
|
3144 |
+
},
|
3145 |
+
{
|
3146 |
+
"epoch": 0.17,
|
3147 |
+
"learning_rate": 0.00018176000000000002,
|
3148 |
+
"loss": 1.9052,
|
3149 |
+
"step": 4560
|
3150 |
+
},
|
3151 |
+
{
|
3152 |
+
"epoch": 0.17,
|
3153 |
+
"learning_rate": 0.00018172,
|
3154 |
+
"loss": 1.9231,
|
3155 |
+
"step": 4570
|
3156 |
+
},
|
3157 |
+
{
|
3158 |
+
"epoch": 0.17,
|
3159 |
+
"learning_rate": 0.00018168,
|
3160 |
+
"loss": 1.9216,
|
3161 |
+
"step": 4580
|
3162 |
+
},
|
3163 |
+
{
|
3164 |
+
"epoch": 0.17,
|
3165 |
+
"learning_rate": 0.00018164,
|
3166 |
+
"loss": 1.8753,
|
3167 |
+
"step": 4590
|
3168 |
+
},
|
3169 |
+
{
|
3170 |
+
"epoch": 0.17,
|
3171 |
+
"learning_rate": 0.00018160000000000002,
|
3172 |
+
"loss": 1.9149,
|
3173 |
+
"step": 4600
|
3174 |
+
},
|
3175 |
+
{
|
3176 |
+
"epoch": 0.17,
|
3177 |
+
"eval_accuracy": 0.6674818619059265,
|
3178 |
+
"eval_loss": 1.6451318264007568,
|
3179 |
+
"eval_runtime": 46.9277,
|
3180 |
+
"eval_samples_per_second": 9.014,
|
3181 |
+
"eval_steps_per_second": 1.129,
|
3182 |
+
"step": 4600
|
3183 |
+
},
|
3184 |
+
{
|
3185 |
+
"epoch": 0.17,
|
3186 |
+
"learning_rate": 0.00018156000000000002,
|
3187 |
+
"loss": 1.8871,
|
3188 |
+
"step": 4610
|
3189 |
+
},
|
3190 |
+
{
|
3191 |
+
"epoch": 0.17,
|
3192 |
+
"learning_rate": 0.00018152,
|
3193 |
+
"loss": 1.9246,
|
3194 |
+
"step": 4620
|
3195 |
+
},
|
3196 |
+
{
|
3197 |
+
"epoch": 0.17,
|
3198 |
+
"learning_rate": 0.00018148,
|
3199 |
+
"loss": 1.9086,
|
3200 |
+
"step": 4630
|
3201 |
+
},
|
3202 |
+
{
|
3203 |
+
"epoch": 0.17,
|
3204 |
+
"learning_rate": 0.00018144,
|
3205 |
+
"loss": 1.9105,
|
3206 |
+
"step": 4640
|
3207 |
+
},
|
3208 |
+
{
|
3209 |
+
"epoch": 0.17,
|
3210 |
+
"learning_rate": 0.00018140000000000002,
|
3211 |
+
"loss": 1.9686,
|
3212 |
+
"step": 4650
|
3213 |
+
},
|
3214 |
+
{
|
3215 |
+
"epoch": 0.17,
|
3216 |
+
"learning_rate": 0.00018136,
|
3217 |
+
"loss": 1.8944,
|
3218 |
+
"step": 4660
|
3219 |
+
},
|
3220 |
+
{
|
3221 |
+
"epoch": 0.17,
|
3222 |
+
"learning_rate": 0.00018132,
|
3223 |
+
"loss": 1.8953,
|
3224 |
+
"step": 4670
|
3225 |
+
},
|
3226 |
+
{
|
3227 |
+
"epoch": 0.17,
|
3228 |
+
"learning_rate": 0.00018128,
|
3229 |
+
"loss": 1.898,
|
3230 |
+
"step": 4680
|
3231 |
+
},
|
3232 |
+
{
|
3233 |
+
"epoch": 0.17,
|
3234 |
+
"learning_rate": 0.00018124000000000002,
|
3235 |
+
"loss": 1.934,
|
3236 |
+
"step": 4690
|
3237 |
+
},
|
3238 |
+
{
|
3239 |
+
"epoch": 0.17,
|
3240 |
+
"learning_rate": 0.0001812,
|
3241 |
+
"loss": 1.8868,
|
3242 |
+
"step": 4700
|
3243 |
+
},
|
3244 |
+
{
|
3245 |
+
"epoch": 0.17,
|
3246 |
+
"eval_accuracy": 0.6653623542838509,
|
3247 |
+
"eval_loss": 1.6440030336380005,
|
3248 |
+
"eval_runtime": 47.0909,
|
3249 |
+
"eval_samples_per_second": 8.983,
|
3250 |
+
"eval_steps_per_second": 1.125,
|
3251 |
+
"step": 4700
|
3252 |
+
},
|
3253 |
+
{
|
3254 |
+
"epoch": 0.17,
|
3255 |
+
"learning_rate": 0.00018116,
|
3256 |
+
"loss": 1.8573,
|
3257 |
+
"step": 4710
|
3258 |
+
},
|
3259 |
+
{
|
3260 |
+
"epoch": 0.17,
|
3261 |
+
"learning_rate": 0.00018112,
|
3262 |
+
"loss": 1.8907,
|
3263 |
+
"step": 4720
|
3264 |
+
},
|
3265 |
+
{
|
3266 |
+
"epoch": 0.17,
|
3267 |
+
"learning_rate": 0.00018108,
|
3268 |
+
"loss": 1.8429,
|
3269 |
+
"step": 4730
|
3270 |
+
},
|
3271 |
+
{
|
3272 |
+
"epoch": 0.18,
|
3273 |
+
"learning_rate": 0.00018104000000000001,
|
3274 |
+
"loss": 1.8393,
|
3275 |
+
"step": 4740
|
3276 |
+
},
|
3277 |
+
{
|
3278 |
+
"epoch": 0.18,
|
3279 |
+
"learning_rate": 0.000181,
|
3280 |
+
"loss": 2.0048,
|
3281 |
+
"step": 4750
|
3282 |
+
},
|
3283 |
+
{
|
3284 |
+
"epoch": 0.18,
|
3285 |
+
"learning_rate": 0.00018096000000000003,
|
3286 |
+
"loss": 1.8358,
|
3287 |
+
"step": 4760
|
3288 |
+
},
|
3289 |
+
{
|
3290 |
+
"epoch": 0.18,
|
3291 |
+
"learning_rate": 0.00018092,
|
3292 |
+
"loss": 1.9148,
|
3293 |
+
"step": 4770
|
3294 |
+
},
|
3295 |
+
{
|
3296 |
+
"epoch": 0.18,
|
3297 |
+
"learning_rate": 0.00018088000000000002,
|
3298 |
+
"loss": 1.8835,
|
3299 |
+
"step": 4780
|
3300 |
+
},
|
3301 |
+
{
|
3302 |
+
"epoch": 0.18,
|
3303 |
+
"learning_rate": 0.00018084,
|
3304 |
+
"loss": 1.8832,
|
3305 |
+
"step": 4790
|
3306 |
+
},
|
3307 |
+
{
|
3308 |
+
"epoch": 0.18,
|
3309 |
+
"learning_rate": 0.0001808,
|
3310 |
+
"loss": 1.9157,
|
3311 |
+
"step": 4800
|
3312 |
+
},
|
3313 |
+
{
|
3314 |
+
"epoch": 0.18,
|
3315 |
+
"eval_accuracy": 0.6643841199967392,
|
3316 |
+
"eval_loss": 1.6713789701461792,
|
3317 |
+
"eval_runtime": 47.3501,
|
3318 |
+
"eval_samples_per_second": 8.933,
|
3319 |
+
"eval_steps_per_second": 1.119,
|
3320 |
+
"step": 4800
|
3321 |
+
},
|
3322 |
+
{
|
3323 |
+
"epoch": 0.18,
|
3324 |
+
"learning_rate": 0.00018076000000000002,
|
3325 |
+
"loss": 1.9131,
|
3326 |
+
"step": 4810
|
3327 |
+
},
|
3328 |
+
{
|
3329 |
+
"epoch": 0.18,
|
3330 |
+
"learning_rate": 0.00018072,
|
3331 |
+
"loss": 1.8841,
|
3332 |
+
"step": 4820
|
3333 |
+
},
|
3334 |
+
{
|
3335 |
+
"epoch": 0.18,
|
3336 |
+
"learning_rate": 0.00018068,
|
3337 |
+
"loss": 1.9479,
|
3338 |
+
"step": 4830
|
3339 |
+
},
|
3340 |
+
{
|
3341 |
+
"epoch": 0.18,
|
3342 |
+
"learning_rate": 0.00018064,
|
3343 |
+
"loss": 1.9109,
|
3344 |
+
"step": 4840
|
3345 |
+
},
|
3346 |
+
{
|
3347 |
+
"epoch": 0.18,
|
3348 |
+
"learning_rate": 0.00018060000000000003,
|
3349 |
+
"loss": 1.9122,
|
3350 |
+
"step": 4850
|
3351 |
+
},
|
3352 |
+
{
|
3353 |
+
"epoch": 0.18,
|
3354 |
+
"learning_rate": 0.00018056000000000002,
|
3355 |
+
"loss": 1.867,
|
3356 |
+
"step": 4860
|
3357 |
+
},
|
3358 |
+
{
|
3359 |
+
"epoch": 0.18,
|
3360 |
+
"learning_rate": 0.00018052,
|
3361 |
+
"loss": 1.9297,
|
3362 |
+
"step": 4870
|
3363 |
+
},
|
3364 |
+
{
|
3365 |
+
"epoch": 0.18,
|
3366 |
+
"learning_rate": 0.00018048,
|
3367 |
+
"loss": 1.8987,
|
3368 |
+
"step": 4880
|
3369 |
+
},
|
3370 |
+
{
|
3371 |
+
"epoch": 0.18,
|
3372 |
+
"learning_rate": 0.00018044,
|
3373 |
+
"loss": 1.8791,
|
3374 |
+
"step": 4890
|
3375 |
+
},
|
3376 |
+
{
|
3377 |
+
"epoch": 0.18,
|
3378 |
+
"learning_rate": 0.00018040000000000002,
|
3379 |
+
"loss": 1.884,
|
3380 |
+
"step": 4900
|
3381 |
+
},
|
3382 |
+
{
|
3383 |
+
"epoch": 0.18,
|
3384 |
+
"eval_accuracy": 0.669764408575854,
|
3385 |
+
"eval_loss": 1.6372376680374146,
|
3386 |
+
"eval_runtime": 45.8594,
|
3387 |
+
"eval_samples_per_second": 9.224,
|
3388 |
+
"eval_steps_per_second": 1.156,
|
3389 |
+
"step": 4900
|
3390 |
+
},
|
3391 |
+
{
|
3392 |
+
"epoch": 0.18,
|
3393 |
+
"learning_rate": 0.00018036000000000001,
|
3394 |
+
"loss": 1.9257,
|
3395 |
+
"step": 4910
|
3396 |
+
},
|
3397 |
+
{
|
3398 |
+
"epoch": 0.18,
|
3399 |
+
"learning_rate": 0.00018032,
|
3400 |
+
"loss": 1.8184,
|
3401 |
+
"step": 4920
|
3402 |
+
},
|
3403 |
+
{
|
3404 |
+
"epoch": 0.18,
|
3405 |
+
"learning_rate": 0.00018028,
|
3406 |
+
"loss": 1.9063,
|
3407 |
+
"step": 4930
|
3408 |
+
},
|
3409 |
+
{
|
3410 |
+
"epoch": 0.18,
|
3411 |
+
"learning_rate": 0.00018024,
|
3412 |
+
"loss": 1.9058,
|
3413 |
+
"step": 4940
|
3414 |
+
},
|
3415 |
+
{
|
3416 |
+
"epoch": 0.18,
|
3417 |
+
"learning_rate": 0.00018020000000000002,
|
3418 |
+
"loss": 1.8876,
|
3419 |
+
"step": 4950
|
3420 |
+
},
|
3421 |
+
{
|
3422 |
+
"epoch": 0.18,
|
3423 |
+
"learning_rate": 0.00018016,
|
3424 |
+
"loss": 1.8873,
|
3425 |
+
"step": 4960
|
3426 |
+
},
|
3427 |
+
{
|
3428 |
+
"epoch": 0.18,
|
3429 |
+
"learning_rate": 0.00018012,
|
3430 |
+
"loss": 1.8623,
|
3431 |
+
"step": 4970
|
3432 |
+
},
|
3433 |
+
{
|
3434 |
+
"epoch": 0.18,
|
3435 |
+
"learning_rate": 0.00018008,
|
3436 |
+
"loss": 1.8649,
|
3437 |
+
"step": 4980
|
3438 |
+
},
|
3439 |
+
{
|
3440 |
+
"epoch": 0.18,
|
3441 |
+
"learning_rate": 0.00018004000000000002,
|
3442 |
+
"loss": 1.9113,
|
3443 |
+
"step": 4990
|
3444 |
+
},
|
3445 |
+
{
|
3446 |
+
"epoch": 0.18,
|
3447 |
+
"learning_rate": 0.00018,
|
3448 |
+
"loss": 1.861,
|
3449 |
+
"step": 5000
|
3450 |
+
},
|
3451 |
+
{
|
3452 |
+
"epoch": 0.18,
|
3453 |
+
"eval_accuracy": 0.6632428466617755,
|
3454 |
+
"eval_loss": 1.6630749702453613,
|
3455 |
+
"eval_runtime": 46.3295,
|
3456 |
+
"eval_samples_per_second": 9.13,
|
3457 |
+
"eval_steps_per_second": 1.144,
|
3458 |
+
"step": 5000
|
3459 |
+
},
|
3460 |
+
{
|
3461 |
+
"epoch": 0.19,
|
3462 |
+
"learning_rate": 0.00017996,
|
3463 |
+
"loss": 1.95,
|
3464 |
+
"step": 5010
|
3465 |
+
},
|
3466 |
+
{
|
3467 |
+
"epoch": 0.19,
|
3468 |
+
"learning_rate": 0.00017992,
|
3469 |
+
"loss": 1.8517,
|
3470 |
+
"step": 5020
|
3471 |
+
},
|
3472 |
+
{
|
3473 |
+
"epoch": 0.19,
|
3474 |
+
"learning_rate": 0.00017988,
|
3475 |
+
"loss": 1.8582,
|
3476 |
+
"step": 5030
|
3477 |
+
},
|
3478 |
+
{
|
3479 |
+
"epoch": 0.19,
|
3480 |
+
"learning_rate": 0.00017984,
|
3481 |
+
"loss": 1.8658,
|
3482 |
+
"step": 5040
|
3483 |
+
},
|
3484 |
+
{
|
3485 |
+
"epoch": 0.19,
|
3486 |
+
"learning_rate": 0.0001798,
|
3487 |
+
"loss": 1.9111,
|
3488 |
+
"step": 5050
|
3489 |
+
},
|
3490 |
+
{
|
3491 |
+
"epoch": 0.19,
|
3492 |
+
"learning_rate": 0.00017976000000000003,
|
3493 |
+
"loss": 1.8874,
|
3494 |
+
"step": 5060
|
3495 |
+
},
|
3496 |
+
{
|
3497 |
+
"epoch": 0.19,
|
3498 |
+
"learning_rate": 0.00017972,
|
3499 |
+
"loss": 1.8853,
|
3500 |
+
"step": 5070
|
3501 |
+
},
|
3502 |
+
{
|
3503 |
+
"epoch": 0.19,
|
3504 |
+
"learning_rate": 0.00017968000000000001,
|
3505 |
+
"loss": 1.8797,
|
3506 |
+
"step": 5080
|
3507 |
+
},
|
3508 |
+
{
|
3509 |
+
"epoch": 0.19,
|
3510 |
+
"learning_rate": 0.00017964,
|
3511 |
+
"loss": 1.8727,
|
3512 |
+
"step": 5090
|
3513 |
+
},
|
3514 |
+
{
|
3515 |
+
"epoch": 0.19,
|
3516 |
+
"learning_rate": 0.0001796,
|
3517 |
+
"loss": 1.9064,
|
3518 |
+
"step": 5100
|
3519 |
+
},
|
3520 |
+
{
|
3521 |
+
"epoch": 0.19,
|
3522 |
+
"eval_accuracy": 0.6621015733268117,
|
3523 |
+
"eval_loss": 1.689902424812317,
|
3524 |
+
"eval_runtime": 47.432,
|
3525 |
+
"eval_samples_per_second": 8.918,
|
3526 |
+
"eval_steps_per_second": 1.117,
|
3527 |
+
"step": 5100
|
3528 |
+
},
|
3529 |
+
{
|
3530 |
+
"epoch": 0.19,
|
3531 |
+
"learning_rate": 0.00017956000000000002,
|
3532 |
+
"loss": 1.8958,
|
3533 |
+
"step": 5110
|
3534 |
+
},
|
3535 |
+
{
|
3536 |
+
"epoch": 0.19,
|
3537 |
+
"learning_rate": 0.00017952,
|
3538 |
+
"loss": 1.8912,
|
3539 |
+
"step": 5120
|
3540 |
+
},
|
3541 |
+
{
|
3542 |
+
"epoch": 0.19,
|
3543 |
+
"learning_rate": 0.00017948,
|
3544 |
+
"loss": 1.8342,
|
3545 |
+
"step": 5130
|
3546 |
+
},
|
3547 |
+
{
|
3548 |
+
"epoch": 0.19,
|
3549 |
+
"learning_rate": 0.00017944,
|
3550 |
+
"loss": 1.864,
|
3551 |
+
"step": 5140
|
3552 |
+
},
|
3553 |
+
{
|
3554 |
+
"epoch": 0.19,
|
3555 |
+
"learning_rate": 0.00017940000000000002,
|
3556 |
+
"loss": 1.8842,
|
3557 |
+
"step": 5150
|
3558 |
+
},
|
3559 |
+
{
|
3560 |
+
"epoch": 0.19,
|
3561 |
+
"learning_rate": 0.00017936000000000002,
|
3562 |
+
"loss": 1.8981,
|
3563 |
+
"step": 5160
|
3564 |
+
},
|
3565 |
+
{
|
3566 |
+
"epoch": 0.19,
|
3567 |
+
"learning_rate": 0.00017932,
|
3568 |
+
"loss": 1.8746,
|
3569 |
+
"step": 5170
|
3570 |
+
},
|
3571 |
+
{
|
3572 |
+
"epoch": 0.19,
|
3573 |
+
"learning_rate": 0.00017928,
|
3574 |
+
"loss": 1.8151,
|
3575 |
+
"step": 5180
|
3576 |
+
},
|
3577 |
+
{
|
3578 |
+
"epoch": 0.19,
|
3579 |
+
"learning_rate": 0.00017924,
|
3580 |
+
"loss": 1.8965,
|
3581 |
+
"step": 5190
|
3582 |
+
},
|
3583 |
+
{
|
3584 |
+
"epoch": 0.19,
|
3585 |
+
"learning_rate": 0.00017920000000000002,
|
3586 |
+
"loss": 1.9178,
|
3587 |
+
"step": 5200
|
3588 |
+
},
|
3589 |
+
{
|
3590 |
+
"epoch": 0.19,
|
3591 |
+
"eval_accuracy": 0.669682889051928,
|
3592 |
+
"eval_loss": 1.654188632965088,
|
3593 |
+
"eval_runtime": 47.2488,
|
3594 |
+
"eval_samples_per_second": 8.953,
|
3595 |
+
"eval_steps_per_second": 1.122,
|
3596 |
+
"step": 5200
|
3597 |
+
},
|
3598 |
+
{
|
3599 |
+
"epoch": 0.19,
|
3600 |
+
"learning_rate": 0.00017916,
|
3601 |
+
"loss": 1.8615,
|
3602 |
+
"step": 5210
|
3603 |
+
},
|
3604 |
+
{
|
3605 |
+
"epoch": 0.19,
|
3606 |
+
"learning_rate": 0.00017912,
|
3607 |
+
"loss": 1.8374,
|
3608 |
+
"step": 5220
|
3609 |
+
},
|
3610 |
+
{
|
3611 |
+
"epoch": 0.19,
|
3612 |
+
"learning_rate": 0.00017908,
|
3613 |
+
"loss": 1.9621,
|
3614 |
+
"step": 5230
|
3615 |
+
},
|
3616 |
+
{
|
3617 |
+
"epoch": 0.19,
|
3618 |
+
"learning_rate": 0.00017904000000000002,
|
3619 |
+
"loss": 1.8499,
|
3620 |
+
"step": 5240
|
3621 |
+
},
|
3622 |
+
{
|
3623 |
+
"epoch": 0.19,
|
3624 |
+
"learning_rate": 0.00017900000000000001,
|
3625 |
+
"loss": 1.8441,
|
3626 |
+
"step": 5250
|
3627 |
+
},
|
3628 |
+
{
|
3629 |
+
"epoch": 0.19,
|
3630 |
+
"learning_rate": 0.00017896,
|
3631 |
+
"loss": 1.9126,
|
3632 |
+
"step": 5260
|
3633 |
+
},
|
3634 |
+
{
|
3635 |
+
"epoch": 0.19,
|
3636 |
+
"learning_rate": 0.00017892,
|
3637 |
+
"loss": 1.874,
|
3638 |
+
"step": 5270
|
3639 |
+
},
|
3640 |
+
{
|
3641 |
+
"epoch": 0.2,
|
3642 |
+
"learning_rate": 0.00017888,
|
3643 |
+
"loss": 1.9262,
|
3644 |
+
"step": 5280
|
3645 |
+
},
|
3646 |
+
{
|
3647 |
+
"epoch": 0.2,
|
3648 |
+
"learning_rate": 0.00017884000000000002,
|
3649 |
+
"loss": 1.8436,
|
3650 |
+
"step": 5290
|
3651 |
+
},
|
3652 |
+
{
|
3653 |
+
"epoch": 0.2,
|
3654 |
+
"learning_rate": 0.0001788,
|
3655 |
+
"loss": 1.9086,
|
3656 |
+
"step": 5300
|
3657 |
+
},
|
3658 |
+
{
|
3659 |
+
"epoch": 0.2,
|
3660 |
+
"eval_accuracy": 0.6702535257194098,
|
3661 |
+
"eval_loss": 1.6423653364181519,
|
3662 |
+
"eval_runtime": 47.3319,
|
3663 |
+
"eval_samples_per_second": 8.937,
|
3664 |
+
"eval_steps_per_second": 1.12,
|
3665 |
+
"step": 5300
|
3666 |
+
},
|
3667 |
+
{
|
3668 |
+
"epoch": 0.2,
|
3669 |
+
"learning_rate": 0.00017876,
|
3670 |
+
"loss": 1.9283,
|
3671 |
+
"step": 5310
|
3672 |
+
},
|
3673 |
+
{
|
3674 |
+
"epoch": 0.2,
|
3675 |
+
"learning_rate": 0.00017872,
|
3676 |
+
"loss": 1.8879,
|
3677 |
+
"step": 5320
|
3678 |
+
},
|
3679 |
+
{
|
3680 |
+
"epoch": 0.2,
|
3681 |
+
"learning_rate": 0.00017868,
|
3682 |
+
"loss": 1.9249,
|
3683 |
+
"step": 5330
|
3684 |
+
},
|
3685 |
+
{
|
3686 |
+
"epoch": 0.2,
|
3687 |
+
"learning_rate": 0.00017864,
|
3688 |
+
"loss": 1.866,
|
3689 |
+
"step": 5340
|
3690 |
+
},
|
3691 |
+
{
|
3692 |
+
"epoch": 0.2,
|
3693 |
+
"learning_rate": 0.0001786,
|
3694 |
+
"loss": 1.8631,
|
3695 |
+
"step": 5350
|
3696 |
+
},
|
3697 |
+
{
|
3698 |
+
"epoch": 0.2,
|
3699 |
+
"learning_rate": 0.00017856000000000003,
|
3700 |
+
"loss": 1.8883,
|
3701 |
+
"step": 5360
|
3702 |
+
},
|
3703 |
+
{
|
3704 |
+
"epoch": 0.2,
|
3705 |
+
"learning_rate": 0.00017852,
|
3706 |
+
"loss": 1.9085,
|
3707 |
+
"step": 5370
|
3708 |
+
},
|
3709 |
+
{
|
3710 |
+
"epoch": 0.2,
|
3711 |
+
"learning_rate": 0.00017848,
|
3712 |
+
"loss": 1.9506,
|
3713 |
+
"step": 5380
|
3714 |
+
},
|
3715 |
+
{
|
3716 |
+
"epoch": 0.2,
|
3717 |
+
"learning_rate": 0.00017844,
|
3718 |
+
"loss": 1.894,
|
3719 |
+
"step": 5390
|
3720 |
+
},
|
3721 |
+
{
|
3722 |
+
"epoch": 0.2,
|
3723 |
+
"learning_rate": 0.0001784,
|
3724 |
+
"loss": 1.9128,
|
3725 |
+
"step": 5400
|
3726 |
+
},
|
3727 |
+
{
|
3728 |
+
"epoch": 0.2,
|
3729 |
+
"eval_accuracy": 0.6704165647672617,
|
3730 |
+
"eval_loss": 1.65791916847229,
|
3731 |
+
"eval_runtime": 46.7984,
|
3732 |
+
"eval_samples_per_second": 9.039,
|
3733 |
+
"eval_steps_per_second": 1.133,
|
3734 |
+
"step": 5400
|
3735 |
+
},
|
3736 |
+
{
|
3737 |
+
"epoch": 0.2,
|
3738 |
+
"learning_rate": 0.00017836000000000002,
|
3739 |
+
"loss": 1.8842,
|
3740 |
+
"step": 5410
|
3741 |
+
},
|
3742 |
+
{
|
3743 |
+
"epoch": 0.2,
|
3744 |
+
"learning_rate": 0.00017832,
|
3745 |
+
"loss": 1.8935,
|
3746 |
+
"step": 5420
|
3747 |
+
},
|
3748 |
+
{
|
3749 |
+
"epoch": 0.2,
|
3750 |
+
"learning_rate": 0.00017828,
|
3751 |
+
"loss": 1.84,
|
3752 |
+
"step": 5430
|
3753 |
+
},
|
3754 |
+
{
|
3755 |
+
"epoch": 0.2,
|
3756 |
+
"learning_rate": 0.00017824,
|
3757 |
+
"loss": 1.8379,
|
3758 |
+
"step": 5440
|
3759 |
+
},
|
3760 |
+
{
|
3761 |
+
"epoch": 0.2,
|
3762 |
+
"learning_rate": 0.00017820000000000002,
|
3763 |
+
"loss": 1.8661,
|
3764 |
+
"step": 5450
|
3765 |
+
},
|
3766 |
+
{
|
3767 |
+
"epoch": 0.2,
|
3768 |
+
"learning_rate": 0.00017816000000000002,
|
3769 |
+
"loss": 1.8418,
|
3770 |
+
"step": 5460
|
3771 |
+
},
|
3772 |
+
{
|
3773 |
+
"epoch": 0.2,
|
3774 |
+
"learning_rate": 0.00017812,
|
3775 |
+
"loss": 1.8773,
|
3776 |
+
"step": 5470
|
3777 |
+
},
|
3778 |
+
{
|
3779 |
+
"epoch": 0.2,
|
3780 |
+
"learning_rate": 0.00017808,
|
3781 |
+
"loss": 1.939,
|
3782 |
+
"step": 5480
|
3783 |
+
},
|
3784 |
+
{
|
3785 |
+
"epoch": 0.2,
|
3786 |
+
"learning_rate": 0.00017804,
|
3787 |
+
"loss": 1.911,
|
3788 |
+
"step": 5490
|
3789 |
+
},
|
3790 |
+
{
|
3791 |
+
"epoch": 0.2,
|
3792 |
+
"learning_rate": 0.00017800000000000002,
|
3793 |
+
"loss": 1.8751,
|
3794 |
+
"step": 5500
|
3795 |
+
},
|
3796 |
+
{
|
3797 |
+
"epoch": 0.2,
|
3798 |
+
"eval_accuracy": 0.6681340180973343,
|
3799 |
+
"eval_loss": 1.6451387405395508,
|
3800 |
+
"eval_runtime": 46.943,
|
3801 |
+
"eval_samples_per_second": 9.011,
|
3802 |
+
"eval_steps_per_second": 1.129,
|
3803 |
+
"step": 5500
|
3804 |
+
},
|
3805 |
+
{
|
3806 |
+
"epoch": 0.2,
|
3807 |
+
"learning_rate": 0.00017796,
|
3808 |
+
"loss": 1.8385,
|
3809 |
+
"step": 5510
|
3810 |
+
},
|
3811 |
+
{
|
3812 |
+
"epoch": 0.2,
|
3813 |
+
"learning_rate": 0.00017792,
|
3814 |
+
"loss": 1.9321,
|
3815 |
+
"step": 5520
|
3816 |
+
},
|
3817 |
+
{
|
3818 |
+
"epoch": 0.2,
|
3819 |
+
"learning_rate": 0.00017788,
|
3820 |
+
"loss": 1.8533,
|
3821 |
+
"step": 5530
|
3822 |
+
},
|
3823 |
+
{
|
3824 |
+
"epoch": 0.2,
|
3825 |
+
"learning_rate": 0.00017784000000000002,
|
3826 |
+
"loss": 1.8428,
|
3827 |
+
"step": 5540
|
3828 |
+
},
|
3829 |
+
{
|
3830 |
+
"epoch": 0.21,
|
3831 |
+
"learning_rate": 0.0001778,
|
3832 |
+
"loss": 1.8833,
|
3833 |
+
"step": 5550
|
3834 |
+
},
|
3835 |
+
{
|
3836 |
+
"epoch": 0.21,
|
3837 |
+
"learning_rate": 0.00017776,
|
3838 |
+
"loss": 1.8387,
|
3839 |
+
"step": 5560
|
3840 |
+
},
|
3841 |
+
{
|
3842 |
+
"epoch": 0.21,
|
3843 |
+
"learning_rate": 0.00017772,
|
3844 |
+
"loss": 1.8742,
|
3845 |
+
"step": 5570
|
3846 |
+
},
|
3847 |
+
{
|
3848 |
+
"epoch": 0.21,
|
3849 |
+
"learning_rate": 0.00017768,
|
3850 |
+
"loss": 1.8835,
|
3851 |
+
"step": 5580
|
3852 |
+
},
|
3853 |
+
{
|
3854 |
+
"epoch": 0.21,
|
3855 |
+
"learning_rate": 0.00017764000000000001,
|
3856 |
+
"loss": 1.9084,
|
3857 |
+
"step": 5590
|
3858 |
+
},
|
3859 |
+
{
|
3860 |
+
"epoch": 0.21,
|
3861 |
+
"learning_rate": 0.0001776,
|
3862 |
+
"loss": 1.8289,
|
3863 |
+
"step": 5600
|
3864 |
+
},
|
3865 |
+
{
|
3866 |
+
"epoch": 0.21,
|
3867 |
+
"eval_accuracy": 0.6712317600065215,
|
3868 |
+
"eval_loss": 1.649842619895935,
|
3869 |
+
"eval_runtime": 47.4408,
|
3870 |
+
"eval_samples_per_second": 8.916,
|
3871 |
+
"eval_steps_per_second": 1.117,
|
3872 |
+
"step": 5600
|
3873 |
+
},
|
3874 |
+
{
|
3875 |
+
"epoch": 0.21,
|
3876 |
+
"learning_rate": 0.00017756000000000003,
|
3877 |
+
"loss": 1.9069,
|
3878 |
+
"step": 5610
|
3879 |
+
},
|
3880 |
+
{
|
3881 |
+
"epoch": 0.21,
|
3882 |
+
"learning_rate": 0.00017752,
|
3883 |
+
"loss": 1.8689,
|
3884 |
+
"step": 5620
|
3885 |
+
},
|
3886 |
+
{
|
3887 |
+
"epoch": 0.21,
|
3888 |
+
"learning_rate": 0.00017748000000000002,
|
3889 |
+
"loss": 1.8327,
|
3890 |
+
"step": 5630
|
3891 |
+
},
|
3892 |
+
{
|
3893 |
+
"epoch": 0.21,
|
3894 |
+
"learning_rate": 0.00017744,
|
3895 |
+
"loss": 1.8061,
|
3896 |
+
"step": 5640
|
3897 |
+
},
|
3898 |
+
{
|
3899 |
+
"epoch": 0.21,
|
3900 |
+
"learning_rate": 0.0001774,
|
3901 |
+
"loss": 1.8382,
|
3902 |
+
"step": 5650
|
3903 |
+
},
|
3904 |
+
{
|
3905 |
+
"epoch": 0.21,
|
3906 |
+
"learning_rate": 0.00017736000000000002,
|
3907 |
+
"loss": 1.8691,
|
3908 |
+
"step": 5660
|
3909 |
+
},
|
3910 |
+
{
|
3911 |
+
"epoch": 0.21,
|
3912 |
+
"learning_rate": 0.00017732000000000002,
|
3913 |
+
"loss": 1.8348,
|
3914 |
+
"step": 5670
|
3915 |
+
},
|
3916 |
+
{
|
3917 |
+
"epoch": 0.21,
|
3918 |
+
"learning_rate": 0.00017728,
|
3919 |
+
"loss": 1.8645,
|
3920 |
+
"step": 5680
|
3921 |
+
},
|
3922 |
+
{
|
3923 |
+
"epoch": 0.21,
|
3924 |
+
"learning_rate": 0.00017724,
|
3925 |
+
"loss": 1.9094,
|
3926 |
+
"step": 5690
|
3927 |
+
},
|
3928 |
+
{
|
3929 |
+
"epoch": 0.21,
|
3930 |
+
"learning_rate": 0.0001772,
|
3931 |
+
"loss": 1.8597,
|
3932 |
+
"step": 5700
|
3933 |
+
},
|
3934 |
+
{
|
3935 |
+
"epoch": 0.21,
|
3936 |
+
"eval_accuracy": 0.669601369528002,
|
3937 |
+
"eval_loss": 1.6287590265274048,
|
3938 |
+
"eval_runtime": 46.725,
|
3939 |
+
"eval_samples_per_second": 9.053,
|
3940 |
+
"eval_steps_per_second": 1.134,
|
3941 |
+
"step": 5700
|
3942 |
+
},
|
3943 |
+
{
|
3944 |
+
"epoch": 0.21,
|
3945 |
+
"learning_rate": 0.00017716000000000002,
|
3946 |
+
"loss": 1.8063,
|
3947 |
+
"step": 5710
|
3948 |
+
},
|
3949 |
+
{
|
3950 |
+
"epoch": 0.21,
|
3951 |
+
"learning_rate": 0.00017712,
|
3952 |
+
"loss": 1.8771,
|
3953 |
+
"step": 5720
|
3954 |
+
},
|
3955 |
+
{
|
3956 |
+
"epoch": 0.21,
|
3957 |
+
"learning_rate": 0.00017708,
|
3958 |
+
"loss": 1.9203,
|
3959 |
+
"step": 5730
|
3960 |
+
},
|
3961 |
+
{
|
3962 |
+
"epoch": 0.21,
|
3963 |
+
"learning_rate": 0.00017704,
|
3964 |
+
"loss": 1.9039,
|
3965 |
+
"step": 5740
|
3966 |
+
},
|
3967 |
+
{
|
3968 |
+
"epoch": 0.21,
|
3969 |
+
"learning_rate": 0.00017700000000000002,
|
3970 |
+
"loss": 1.8713,
|
3971 |
+
"step": 5750
|
3972 |
+
},
|
3973 |
+
{
|
3974 |
+
"epoch": 0.21,
|
3975 |
+
"learning_rate": 0.00017696,
|
3976 |
+
"loss": 1.8446,
|
3977 |
+
"step": 5760
|
3978 |
+
},
|
3979 |
+
{
|
3980 |
+
"epoch": 0.21,
|
3981 |
+
"learning_rate": 0.00017692,
|
3982 |
+
"loss": 1.8771,
|
3983 |
+
"step": 5770
|
3984 |
+
},
|
3985 |
+
{
|
3986 |
+
"epoch": 0.21,
|
3987 |
+
"learning_rate": 0.00017688,
|
3988 |
+
"loss": 1.8959,
|
3989 |
+
"step": 5780
|
3990 |
+
},
|
3991 |
+
{
|
3992 |
+
"epoch": 0.21,
|
3993 |
+
"learning_rate": 0.00017684,
|
3994 |
+
"loss": 1.915,
|
3995 |
+
"step": 5790
|
3996 |
+
},
|
3997 |
+
{
|
3998 |
+
"epoch": 0.21,
|
3999 |
+
"learning_rate": 0.00017680000000000001,
|
4000 |
+
"loss": 1.8873,
|
4001 |
+
"step": 5800
|
4002 |
+
},
|
4003 |
+
{
|
4004 |
+
"epoch": 0.21,
|
4005 |
+
"eval_accuracy": 0.6718023966740034,
|
4006 |
+
"eval_loss": 1.634265422821045,
|
4007 |
+
"eval_runtime": 47.1163,
|
4008 |
+
"eval_samples_per_second": 8.978,
|
4009 |
+
"eval_steps_per_second": 1.125,
|
4010 |
+
"step": 5800
|
4011 |
+
}
|
4012 |
+
],
|
4013 |
+
"max_steps": 50000,
|
4014 |
+
"num_train_epochs": 2,
|
4015 |
+
"total_flos": 1.066306262532096e+17,
|
4016 |
+
"trial_name": null,
|
4017 |
+
"trial_params": null
|
4018 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0a9cd9cf01a1ff40ca4c7f3b787db9616f0da4cd4bacee79ce247ef18fa96834
|
3 |
+
size 4015
|