ChlorophyllChampion
commited on
Commit
•
975e103
1
Parent(s):
0c627ad
Upload 8 files
Browse files- config.json +29 -0
- generation_config.json +5 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +3921 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"GPTNeoXForCausalLM"
|
4 |
+
],
|
5 |
+
"attention_bias": true,
|
6 |
+
"attention_dropout": 0.0,
|
7 |
+
"bos_token_id": null,
|
8 |
+
"classifier_dropout": 0.1,
|
9 |
+
"eos_token_id": 1,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout": 0.0,
|
12 |
+
"hidden_size": 512,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 2048,
|
15 |
+
"layer_norm_eps": 1e-05,
|
16 |
+
"max_position_embeddings": 2048,
|
17 |
+
"model_type": "gpt_neox",
|
18 |
+
"num_attention_heads": 8,
|
19 |
+
"num_hidden_layers": 6,
|
20 |
+
"rope_scaling": null,
|
21 |
+
"rotary_emb_base": 10000,
|
22 |
+
"rotary_pct": 0.25,
|
23 |
+
"tie_word_embeddings": false,
|
24 |
+
"torch_dtype": "float32",
|
25 |
+
"transformers_version": "4.39.2",
|
26 |
+
"use_cache": true,
|
27 |
+
"use_parallel_residual": true,
|
28 |
+
"vocab_size": 384
|
29 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"eos_token_id": 1,
|
4 |
+
"transformers_version": "4.39.2"
|
5 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cd75b7f8c67702815af97644adc44cb8293ea3d6ab29636c11810aa72fd78a47
|
3 |
+
size 77242672
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b51a5b06a16148881aef4419f115048eae5462c2e56366c6ba1c9a8c9d2c3c5d
|
3 |
+
size 154513210
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:592c43fcd4f224a713b94e9b0547d56fff2d04c9a3f50aaae922858b09cf4f47
|
3 |
+
size 14244
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:54170defda7f8314ce25911dc27a6c89c123f5c8df2ad3e4ecbdadd81aee4d82
|
3 |
+
size 1000
|
trainer_state.json
ADDED
@@ -0,0 +1,3921 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 2.5235376358032227,
|
3 |
+
"best_model_checkpoint": "./results/checkpoint-50000",
|
4 |
+
"epoch": 0.3637362257638006,
|
5 |
+
"eval_steps": 1000,
|
6 |
+
"global_step": 50000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"grad_norm": 3.156810998916626,
|
14 |
+
"learning_rate": 5e-06,
|
15 |
+
"loss": 5.0881,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.0,
|
20 |
+
"grad_norm": 1.8936021327972412,
|
21 |
+
"learning_rate": 1e-05,
|
22 |
+
"loss": 3.4609,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.0,
|
27 |
+
"grad_norm": 0.8277048468589783,
|
28 |
+
"learning_rate": 1.5e-05,
|
29 |
+
"loss": 2.9956,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.0,
|
34 |
+
"grad_norm": 0.5409151315689087,
|
35 |
+
"learning_rate": 2e-05,
|
36 |
+
"loss": 2.7702,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.0,
|
41 |
+
"grad_norm": 0.37298905849456787,
|
42 |
+
"learning_rate": 2.5e-05,
|
43 |
+
"loss": 2.7116,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.0,
|
48 |
+
"grad_norm": 1.4856789112091064,
|
49 |
+
"learning_rate": 3e-05,
|
50 |
+
"loss": 2.676,
|
51 |
+
"step": 600
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.01,
|
55 |
+
"grad_norm": 0.9477183222770691,
|
56 |
+
"learning_rate": 3.5e-05,
|
57 |
+
"loss": 2.6509,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.01,
|
62 |
+
"grad_norm": 0.42317140102386475,
|
63 |
+
"learning_rate": 4e-05,
|
64 |
+
"loss": 2.6364,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.01,
|
69 |
+
"grad_norm": 0.9827988743782043,
|
70 |
+
"learning_rate": 4.5e-05,
|
71 |
+
"loss": 2.6289,
|
72 |
+
"step": 900
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.01,
|
76 |
+
"grad_norm": 0.647011935710907,
|
77 |
+
"learning_rate": 5e-05,
|
78 |
+
"loss": 2.6236,
|
79 |
+
"step": 1000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.01,
|
83 |
+
"eval_loss": 2.6208438873291016,
|
84 |
+
"eval_runtime": 5241.4331,
|
85 |
+
"eval_samples_per_second": 1118.979,
|
86 |
+
"eval_steps_per_second": 69.936,
|
87 |
+
"step": 1000
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.01,
|
91 |
+
"grad_norm": 0.3872681260108948,
|
92 |
+
"learning_rate": 5.500000000000001e-05,
|
93 |
+
"loss": 2.6201,
|
94 |
+
"step": 1100
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.01,
|
98 |
+
"grad_norm": 0.6852623224258423,
|
99 |
+
"learning_rate": 6e-05,
|
100 |
+
"loss": 2.6152,
|
101 |
+
"step": 1200
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.01,
|
105 |
+
"grad_norm": 0.5895536541938782,
|
106 |
+
"learning_rate": 6.500000000000001e-05,
|
107 |
+
"loss": 2.6152,
|
108 |
+
"step": 1300
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 0.01,
|
112 |
+
"grad_norm": 0.5872493386268616,
|
113 |
+
"learning_rate": 7e-05,
|
114 |
+
"loss": 2.6124,
|
115 |
+
"step": 1400
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 0.01,
|
119 |
+
"grad_norm": 0.6140819787979126,
|
120 |
+
"learning_rate": 7.500000000000001e-05,
|
121 |
+
"loss": 2.6106,
|
122 |
+
"step": 1500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.01,
|
126 |
+
"grad_norm": 0.6721034646034241,
|
127 |
+
"learning_rate": 8e-05,
|
128 |
+
"loss": 2.6097,
|
129 |
+
"step": 1600
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.01,
|
133 |
+
"grad_norm": 0.3682301640510559,
|
134 |
+
"learning_rate": 8.5e-05,
|
135 |
+
"loss": 2.6075,
|
136 |
+
"step": 1700
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.01,
|
140 |
+
"grad_norm": 0.4152977764606476,
|
141 |
+
"learning_rate": 9e-05,
|
142 |
+
"loss": 2.6065,
|
143 |
+
"step": 1800
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 0.01,
|
147 |
+
"grad_norm": 0.30114710330963135,
|
148 |
+
"learning_rate": 9.5e-05,
|
149 |
+
"loss": 2.6067,
|
150 |
+
"step": 1900
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.01,
|
154 |
+
"grad_norm": 0.29172569513320923,
|
155 |
+
"learning_rate": 0.0001,
|
156 |
+
"loss": 2.6056,
|
157 |
+
"step": 2000
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 0.01,
|
161 |
+
"eval_loss": 2.604510545730591,
|
162 |
+
"eval_runtime": 5009.0383,
|
163 |
+
"eval_samples_per_second": 1170.894,
|
164 |
+
"eval_steps_per_second": 73.181,
|
165 |
+
"step": 2000
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.02,
|
169 |
+
"grad_norm": 0.26403701305389404,
|
170 |
+
"learning_rate": 9.999999909099618e-05,
|
171 |
+
"loss": 2.6061,
|
172 |
+
"step": 2100
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 0.02,
|
176 |
+
"grad_norm": 0.42730650305747986,
|
177 |
+
"learning_rate": 9.999999636398476e-05,
|
178 |
+
"loss": 2.6038,
|
179 |
+
"step": 2200
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 0.02,
|
183 |
+
"grad_norm": 0.2776353657245636,
|
184 |
+
"learning_rate": 9.999999181896581e-05,
|
185 |
+
"loss": 2.6048,
|
186 |
+
"step": 2300
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 0.02,
|
190 |
+
"grad_norm": 0.2872585952281952,
|
191 |
+
"learning_rate": 9.999998545593951e-05,
|
192 |
+
"loss": 2.6031,
|
193 |
+
"step": 2400
|
194 |
+
},
|
195 |
+
{
|
196 |
+
"epoch": 0.02,
|
197 |
+
"grad_norm": 0.24884726107120514,
|
198 |
+
"learning_rate": 9.999997727490612e-05,
|
199 |
+
"loss": 2.6007,
|
200 |
+
"step": 2500
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.02,
|
204 |
+
"grad_norm": 0.2905128002166748,
|
205 |
+
"learning_rate": 9.99999672758659e-05,
|
206 |
+
"loss": 2.6001,
|
207 |
+
"step": 2600
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 0.02,
|
211 |
+
"grad_norm": 0.22506840527057648,
|
212 |
+
"learning_rate": 9.999995545881924e-05,
|
213 |
+
"loss": 2.6002,
|
214 |
+
"step": 2700
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 0.02,
|
218 |
+
"grad_norm": 0.235252246260643,
|
219 |
+
"learning_rate": 9.999994182376653e-05,
|
220 |
+
"loss": 2.5991,
|
221 |
+
"step": 2800
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.02,
|
225 |
+
"grad_norm": 0.3328251838684082,
|
226 |
+
"learning_rate": 9.999992637070832e-05,
|
227 |
+
"loss": 2.5997,
|
228 |
+
"step": 2900
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 0.02,
|
232 |
+
"grad_norm": 0.23325322568416595,
|
233 |
+
"learning_rate": 9.999990909964513e-05,
|
234 |
+
"loss": 2.5995,
|
235 |
+
"step": 3000
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 0.02,
|
239 |
+
"eval_loss": 2.598254919052124,
|
240 |
+
"eval_runtime": 5203.4796,
|
241 |
+
"eval_samples_per_second": 1127.141,
|
242 |
+
"eval_steps_per_second": 70.446,
|
243 |
+
"step": 3000
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 0.02,
|
247 |
+
"grad_norm": 0.2796269655227661,
|
248 |
+
"learning_rate": 9.99998900105776e-05,
|
249 |
+
"loss": 2.5984,
|
250 |
+
"step": 3100
|
251 |
+
},
|
252 |
+
{
|
253 |
+
"epoch": 0.02,
|
254 |
+
"grad_norm": 0.21093028783798218,
|
255 |
+
"learning_rate": 9.999986910350642e-05,
|
256 |
+
"loss": 2.5985,
|
257 |
+
"step": 3200
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.02,
|
261 |
+
"grad_norm": 0.22558662295341492,
|
262 |
+
"learning_rate": 9.999984637843238e-05,
|
263 |
+
"loss": 2.5978,
|
264 |
+
"step": 3300
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 0.02,
|
268 |
+
"grad_norm": 0.3363204598426819,
|
269 |
+
"learning_rate": 9.999982183535627e-05,
|
270 |
+
"loss": 2.5965,
|
271 |
+
"step": 3400
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 0.03,
|
275 |
+
"grad_norm": 0.27081742882728577,
|
276 |
+
"learning_rate": 9.9999795474279e-05,
|
277 |
+
"loss": 2.5961,
|
278 |
+
"step": 3500
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 0.03,
|
282 |
+
"grad_norm": 0.23502503335475922,
|
283 |
+
"learning_rate": 9.999976729520151e-05,
|
284 |
+
"loss": 2.5956,
|
285 |
+
"step": 3600
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 0.03,
|
289 |
+
"grad_norm": 0.17260690033435822,
|
290 |
+
"learning_rate": 9.999973729812485e-05,
|
291 |
+
"loss": 2.5955,
|
292 |
+
"step": 3700
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 0.03,
|
296 |
+
"grad_norm": 0.16165785491466522,
|
297 |
+
"learning_rate": 9.999970548305009e-05,
|
298 |
+
"loss": 2.5955,
|
299 |
+
"step": 3800
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 0.03,
|
303 |
+
"grad_norm": 0.24890035390853882,
|
304 |
+
"learning_rate": 9.99996718499784e-05,
|
305 |
+
"loss": 2.5946,
|
306 |
+
"step": 3900
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 0.03,
|
310 |
+
"grad_norm": 0.30852416157722473,
|
311 |
+
"learning_rate": 9.999963639891102e-05,
|
312 |
+
"loss": 2.5956,
|
313 |
+
"step": 4000
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 0.03,
|
317 |
+
"eval_loss": 2.594769239425659,
|
318 |
+
"eval_runtime": 5243.3493,
|
319 |
+
"eval_samples_per_second": 1118.57,
|
320 |
+
"eval_steps_per_second": 69.911,
|
321 |
+
"step": 4000
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 0.03,
|
325 |
+
"grad_norm": 0.37085938453674316,
|
326 |
+
"learning_rate": 9.999959912984918e-05,
|
327 |
+
"loss": 2.5947,
|
328 |
+
"step": 4100
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 0.03,
|
332 |
+
"grad_norm": 0.3395659029483795,
|
333 |
+
"learning_rate": 9.999956004279429e-05,
|
334 |
+
"loss": 2.5955,
|
335 |
+
"step": 4200
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 0.03,
|
339 |
+
"grad_norm": 0.16178911924362183,
|
340 |
+
"learning_rate": 9.999951913774777e-05,
|
341 |
+
"loss": 2.594,
|
342 |
+
"step": 4300
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 0.03,
|
346 |
+
"grad_norm": 0.32948923110961914,
|
347 |
+
"learning_rate": 9.999947641471107e-05,
|
348 |
+
"loss": 2.5941,
|
349 |
+
"step": 4400
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 0.03,
|
353 |
+
"grad_norm": 0.20488527417182922,
|
354 |
+
"learning_rate": 9.999943187368577e-05,
|
355 |
+
"loss": 2.5934,
|
356 |
+
"step": 4500
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 0.03,
|
360 |
+
"grad_norm": 0.2330751270055771,
|
361 |
+
"learning_rate": 9.999938551467348e-05,
|
362 |
+
"loss": 2.593,
|
363 |
+
"step": 4600
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 0.03,
|
367 |
+
"grad_norm": 0.2759188711643219,
|
368 |
+
"learning_rate": 9.99993373376759e-05,
|
369 |
+
"loss": 2.5934,
|
370 |
+
"step": 4700
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 0.03,
|
374 |
+
"grad_norm": 0.20582658052444458,
|
375 |
+
"learning_rate": 9.999928734269477e-05,
|
376 |
+
"loss": 2.5921,
|
377 |
+
"step": 4800
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 0.04,
|
381 |
+
"grad_norm": 0.27048927545547485,
|
382 |
+
"learning_rate": 9.99992355297319e-05,
|
383 |
+
"loss": 2.5908,
|
384 |
+
"step": 4900
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 0.04,
|
388 |
+
"grad_norm": 0.23385432362556458,
|
389 |
+
"learning_rate": 9.999918189878918e-05,
|
390 |
+
"loss": 2.5921,
|
391 |
+
"step": 5000
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 0.04,
|
395 |
+
"eval_loss": 2.5919315814971924,
|
396 |
+
"eval_runtime": 5496.6841,
|
397 |
+
"eval_samples_per_second": 1067.017,
|
398 |
+
"eval_steps_per_second": 66.689,
|
399 |
+
"step": 5000
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 0.04,
|
403 |
+
"grad_norm": 0.22557920217514038,
|
404 |
+
"learning_rate": 9.999912644986859e-05,
|
405 |
+
"loss": 2.5931,
|
406 |
+
"step": 5100
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 0.04,
|
410 |
+
"grad_norm": 0.23983053863048553,
|
411 |
+
"learning_rate": 9.99990691829721e-05,
|
412 |
+
"loss": 2.5915,
|
413 |
+
"step": 5200
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 0.04,
|
417 |
+
"grad_norm": 0.18743467330932617,
|
418 |
+
"learning_rate": 9.999901009810181e-05,
|
419 |
+
"loss": 2.5913,
|
420 |
+
"step": 5300
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 0.04,
|
424 |
+
"grad_norm": 0.20421144366264343,
|
425 |
+
"learning_rate": 9.999894919525987e-05,
|
426 |
+
"loss": 2.59,
|
427 |
+
"step": 5400
|
428 |
+
},
|
429 |
+
{
|
430 |
+
"epoch": 0.04,
|
431 |
+
"grad_norm": 0.19018307328224182,
|
432 |
+
"learning_rate": 9.999888647444851e-05,
|
433 |
+
"loss": 2.5908,
|
434 |
+
"step": 5500
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.04,
|
438 |
+
"grad_norm": 0.23669780790805817,
|
439 |
+
"learning_rate": 9.999882193566997e-05,
|
440 |
+
"loss": 2.5908,
|
441 |
+
"step": 5600
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 0.04,
|
445 |
+
"grad_norm": 0.20617301762104034,
|
446 |
+
"learning_rate": 9.999875557892664e-05,
|
447 |
+
"loss": 2.5902,
|
448 |
+
"step": 5700
|
449 |
+
},
|
450 |
+
{
|
451 |
+
"epoch": 0.04,
|
452 |
+
"grad_norm": 0.1931978017091751,
|
453 |
+
"learning_rate": 9.999868740422092e-05,
|
454 |
+
"loss": 2.5907,
|
455 |
+
"step": 5800
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 0.04,
|
459 |
+
"grad_norm": 0.19971776008605957,
|
460 |
+
"learning_rate": 9.999861741155526e-05,
|
461 |
+
"loss": 2.5885,
|
462 |
+
"step": 5900
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 0.04,
|
466 |
+
"grad_norm": 0.21058019995689392,
|
467 |
+
"learning_rate": 9.999854560093225e-05,
|
468 |
+
"loss": 2.5897,
|
469 |
+
"step": 6000
|
470 |
+
},
|
471 |
+
{
|
472 |
+
"epoch": 0.04,
|
473 |
+
"eval_loss": 2.5891263484954834,
|
474 |
+
"eval_runtime": 5309.84,
|
475 |
+
"eval_samples_per_second": 1104.563,
|
476 |
+
"eval_steps_per_second": 69.035,
|
477 |
+
"step": 6000
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 0.04,
|
481 |
+
"grad_norm": 0.2691722512245178,
|
482 |
+
"learning_rate": 9.999847197235446e-05,
|
483 |
+
"loss": 2.5882,
|
484 |
+
"step": 6100
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 0.05,
|
488 |
+
"grad_norm": 0.2411184161901474,
|
489 |
+
"learning_rate": 9.99983965258246e-05,
|
490 |
+
"loss": 2.589,
|
491 |
+
"step": 6200
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 0.05,
|
495 |
+
"grad_norm": 0.15842053294181824,
|
496 |
+
"learning_rate": 9.99983192613454e-05,
|
497 |
+
"loss": 2.5892,
|
498 |
+
"step": 6300
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 0.05,
|
502 |
+
"grad_norm": 0.17729991674423218,
|
503 |
+
"learning_rate": 9.999824017891965e-05,
|
504 |
+
"loss": 2.588,
|
505 |
+
"step": 6400
|
506 |
+
},
|
507 |
+
{
|
508 |
+
"epoch": 0.05,
|
509 |
+
"grad_norm": 0.22472384572029114,
|
510 |
+
"learning_rate": 9.999815927855027e-05,
|
511 |
+
"loss": 2.5873,
|
512 |
+
"step": 6500
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 0.05,
|
516 |
+
"grad_norm": 0.2445155680179596,
|
517 |
+
"learning_rate": 9.999807656024016e-05,
|
518 |
+
"loss": 2.5871,
|
519 |
+
"step": 6600
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 0.05,
|
523 |
+
"grad_norm": 0.2309512048959732,
|
524 |
+
"learning_rate": 9.999799202399236e-05,
|
525 |
+
"loss": 2.5872,
|
526 |
+
"step": 6700
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 0.05,
|
530 |
+
"grad_norm": 0.24274510145187378,
|
531 |
+
"learning_rate": 9.999790566980991e-05,
|
532 |
+
"loss": 2.5863,
|
533 |
+
"step": 6800
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 0.05,
|
537 |
+
"grad_norm": 0.2553771436214447,
|
538 |
+
"learning_rate": 9.999781749769597e-05,
|
539 |
+
"loss": 2.5866,
|
540 |
+
"step": 6900
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 0.05,
|
544 |
+
"grad_norm": 0.2770930230617523,
|
545 |
+
"learning_rate": 9.999772750765375e-05,
|
546 |
+
"loss": 2.5867,
|
547 |
+
"step": 7000
|
548 |
+
},
|
549 |
+
{
|
550 |
+
"epoch": 0.05,
|
551 |
+
"eval_loss": 2.5862584114074707,
|
552 |
+
"eval_runtime": 5428.0291,
|
553 |
+
"eval_samples_per_second": 1080.512,
|
554 |
+
"eval_steps_per_second": 67.532,
|
555 |
+
"step": 7000
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.05,
|
559 |
+
"grad_norm": 0.20760175585746765,
|
560 |
+
"learning_rate": 9.999763569968652e-05,
|
561 |
+
"loss": 2.5864,
|
562 |
+
"step": 7100
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.05,
|
566 |
+
"grad_norm": 0.195896178483963,
|
567 |
+
"learning_rate": 9.999754207379762e-05,
|
568 |
+
"loss": 2.5859,
|
569 |
+
"step": 7200
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.05,
|
573 |
+
"grad_norm": 0.17725147306919098,
|
574 |
+
"learning_rate": 9.999744662999042e-05,
|
575 |
+
"loss": 2.5862,
|
576 |
+
"step": 7300
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.05,
|
580 |
+
"grad_norm": 0.19556549191474915,
|
581 |
+
"learning_rate": 9.999734936826843e-05,
|
582 |
+
"loss": 2.5849,
|
583 |
+
"step": 7400
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.05,
|
587 |
+
"grad_norm": 0.1898818463087082,
|
588 |
+
"learning_rate": 9.999725028863518e-05,
|
589 |
+
"loss": 2.5838,
|
590 |
+
"step": 7500
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.06,
|
594 |
+
"grad_norm": 0.1693861335515976,
|
595 |
+
"learning_rate": 9.999714939109426e-05,
|
596 |
+
"loss": 2.5842,
|
597 |
+
"step": 7600
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.06,
|
601 |
+
"grad_norm": 0.2052980363368988,
|
602 |
+
"learning_rate": 9.999704667564935e-05,
|
603 |
+
"loss": 2.5846,
|
604 |
+
"step": 7700
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.06,
|
608 |
+
"grad_norm": 0.24053046107292175,
|
609 |
+
"learning_rate": 9.999694214230418e-05,
|
610 |
+
"loss": 2.5853,
|
611 |
+
"step": 7800
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.06,
|
615 |
+
"grad_norm": 0.18848931789398193,
|
616 |
+
"learning_rate": 9.999683579106255e-05,
|
617 |
+
"loss": 2.5847,
|
618 |
+
"step": 7900
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.06,
|
622 |
+
"grad_norm": 0.21658702194690704,
|
623 |
+
"learning_rate": 9.999672762192834e-05,
|
624 |
+
"loss": 2.5834,
|
625 |
+
"step": 8000
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.06,
|
629 |
+
"eval_loss": 2.5834951400756836,
|
630 |
+
"eval_runtime": 5812.5431,
|
631 |
+
"eval_samples_per_second": 1009.034,
|
632 |
+
"eval_steps_per_second": 63.065,
|
633 |
+
"step": 8000
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 0.06,
|
637 |
+
"grad_norm": 0.2054041624069214,
|
638 |
+
"learning_rate": 9.999661763490544e-05,
|
639 |
+
"loss": 2.5846,
|
640 |
+
"step": 8100
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 0.06,
|
644 |
+
"grad_norm": 0.19781599938869476,
|
645 |
+
"learning_rate": 9.99965058299979e-05,
|
646 |
+
"loss": 2.5828,
|
647 |
+
"step": 8200
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 0.06,
|
651 |
+
"grad_norm": 0.25480276346206665,
|
652 |
+
"learning_rate": 9.999639220720978e-05,
|
653 |
+
"loss": 2.5824,
|
654 |
+
"step": 8300
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 0.06,
|
658 |
+
"grad_norm": 0.24584674835205078,
|
659 |
+
"learning_rate": 9.999627676654517e-05,
|
660 |
+
"loss": 2.5815,
|
661 |
+
"step": 8400
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 0.06,
|
665 |
+
"grad_norm": 0.21407808363437653,
|
666 |
+
"learning_rate": 9.999615950800832e-05,
|
667 |
+
"loss": 2.5816,
|
668 |
+
"step": 8500
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 0.06,
|
672 |
+
"grad_norm": 0.19315016269683838,
|
673 |
+
"learning_rate": 9.999604043160346e-05,
|
674 |
+
"loss": 2.582,
|
675 |
+
"step": 8600
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 0.06,
|
679 |
+
"grad_norm": 0.18959668278694153,
|
680 |
+
"learning_rate": 9.999591953733491e-05,
|
681 |
+
"loss": 2.5823,
|
682 |
+
"step": 8700
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 0.06,
|
686 |
+
"grad_norm": 0.2016109973192215,
|
687 |
+
"learning_rate": 9.99957968252071e-05,
|
688 |
+
"loss": 2.5817,
|
689 |
+
"step": 8800
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.06,
|
693 |
+
"grad_norm": 0.1687597781419754,
|
694 |
+
"learning_rate": 9.999567229522448e-05,
|
695 |
+
"loss": 2.5808,
|
696 |
+
"step": 8900
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 0.07,
|
700 |
+
"grad_norm": 0.17627790570259094,
|
701 |
+
"learning_rate": 9.999554594739155e-05,
|
702 |
+
"loss": 2.5809,
|
703 |
+
"step": 9000
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 0.07,
|
707 |
+
"eval_loss": 2.5812571048736572,
|
708 |
+
"eval_runtime": 5682.2226,
|
709 |
+
"eval_samples_per_second": 1032.176,
|
710 |
+
"eval_steps_per_second": 64.511,
|
711 |
+
"step": 9000
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 0.07,
|
715 |
+
"grad_norm": 0.12426480650901794,
|
716 |
+
"learning_rate": 9.999541778171295e-05,
|
717 |
+
"loss": 2.5808,
|
718 |
+
"step": 9100
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 0.07,
|
722 |
+
"grad_norm": 0.20493678748607635,
|
723 |
+
"learning_rate": 9.999528779819331e-05,
|
724 |
+
"loss": 2.582,
|
725 |
+
"step": 9200
|
726 |
+
},
|
727 |
+
{
|
728 |
+
"epoch": 0.07,
|
729 |
+
"grad_norm": 0.2577107548713684,
|
730 |
+
"learning_rate": 9.999515599683736e-05,
|
731 |
+
"loss": 2.5821,
|
732 |
+
"step": 9300
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 0.07,
|
736 |
+
"grad_norm": 0.17233897745609283,
|
737 |
+
"learning_rate": 9.999502237764991e-05,
|
738 |
+
"loss": 2.5805,
|
739 |
+
"step": 9400
|
740 |
+
},
|
741 |
+
{
|
742 |
+
"epoch": 0.07,
|
743 |
+
"grad_norm": 0.17032714188098907,
|
744 |
+
"learning_rate": 9.99948869406358e-05,
|
745 |
+
"loss": 2.5797,
|
746 |
+
"step": 9500
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 0.07,
|
750 |
+
"grad_norm": 0.21331337094306946,
|
751 |
+
"learning_rate": 9.999474968579994e-05,
|
752 |
+
"loss": 2.5807,
|
753 |
+
"step": 9600
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 0.07,
|
757 |
+
"grad_norm": 0.1757407933473587,
|
758 |
+
"learning_rate": 9.999461061314734e-05,
|
759 |
+
"loss": 2.5796,
|
760 |
+
"step": 9700
|
761 |
+
},
|
762 |
+
{
|
763 |
+
"epoch": 0.07,
|
764 |
+
"grad_norm": 0.20256169140338898,
|
765 |
+
"learning_rate": 9.999446972268308e-05,
|
766 |
+
"loss": 2.5805,
|
767 |
+
"step": 9800
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 0.07,
|
771 |
+
"grad_norm": 0.16938987374305725,
|
772 |
+
"learning_rate": 9.999432701441223e-05,
|
773 |
+
"loss": 2.5793,
|
774 |
+
"step": 9900
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 0.07,
|
778 |
+
"grad_norm": 0.15477782487869263,
|
779 |
+
"learning_rate": 9.999418248834002e-05,
|
780 |
+
"loss": 2.5797,
|
781 |
+
"step": 10000
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 0.07,
|
785 |
+
"eval_loss": 2.579577684402466,
|
786 |
+
"eval_runtime": 5716.7677,
|
787 |
+
"eval_samples_per_second": 1025.939,
|
788 |
+
"eval_steps_per_second": 64.121,
|
789 |
+
"step": 10000
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 0.07,
|
793 |
+
"grad_norm": 0.1892194151878357,
|
794 |
+
"learning_rate": 9.99940361444717e-05,
|
795 |
+
"loss": 2.5807,
|
796 |
+
"step": 10100
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 0.07,
|
800 |
+
"grad_norm": 0.1950315237045288,
|
801 |
+
"learning_rate": 9.999388798281258e-05,
|
802 |
+
"loss": 2.5803,
|
803 |
+
"step": 10200
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 0.07,
|
807 |
+
"grad_norm": 0.182656928896904,
|
808 |
+
"learning_rate": 9.999373800336806e-05,
|
809 |
+
"loss": 2.579,
|
810 |
+
"step": 10300
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 0.08,
|
814 |
+
"grad_norm": 0.2610529959201813,
|
815 |
+
"learning_rate": 9.999358620614357e-05,
|
816 |
+
"loss": 2.5798,
|
817 |
+
"step": 10400
|
818 |
+
},
|
819 |
+
{
|
820 |
+
"epoch": 0.08,
|
821 |
+
"grad_norm": 0.1903856247663498,
|
822 |
+
"learning_rate": 9.999343259114464e-05,
|
823 |
+
"loss": 2.5782,
|
824 |
+
"step": 10500
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 0.08,
|
828 |
+
"grad_norm": 0.2288079857826233,
|
829 |
+
"learning_rate": 9.999327715837687e-05,
|
830 |
+
"loss": 2.5798,
|
831 |
+
"step": 10600
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 0.08,
|
835 |
+
"grad_norm": 0.22872629761695862,
|
836 |
+
"learning_rate": 9.99931199078459e-05,
|
837 |
+
"loss": 2.5777,
|
838 |
+
"step": 10700
|
839 |
+
},
|
840 |
+
{
|
841 |
+
"epoch": 0.08,
|
842 |
+
"grad_norm": 0.22784186899662018,
|
843 |
+
"learning_rate": 9.999296083955744e-05,
|
844 |
+
"loss": 2.5783,
|
845 |
+
"step": 10800
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 0.08,
|
849 |
+
"grad_norm": 0.1781127005815506,
|
850 |
+
"learning_rate": 9.999279995351729e-05,
|
851 |
+
"loss": 2.5784,
|
852 |
+
"step": 10900
|
853 |
+
},
|
854 |
+
{
|
855 |
+
"epoch": 0.08,
|
856 |
+
"grad_norm": 0.19801098108291626,
|
857 |
+
"learning_rate": 9.99926372497313e-05,
|
858 |
+
"loss": 2.5786,
|
859 |
+
"step": 11000
|
860 |
+
},
|
861 |
+
{
|
862 |
+
"epoch": 0.08,
|
863 |
+
"eval_loss": 2.5776655673980713,
|
864 |
+
"eval_runtime": 5841.1118,
|
865 |
+
"eval_samples_per_second": 1004.099,
|
866 |
+
"eval_steps_per_second": 62.756,
|
867 |
+
"step": 11000
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 0.08,
|
871 |
+
"grad_norm": 0.24839451909065247,
|
872 |
+
"learning_rate": 9.999247272820536e-05,
|
873 |
+
"loss": 2.5777,
|
874 |
+
"step": 11100
|
875 |
+
},
|
876 |
+
{
|
877 |
+
"epoch": 0.08,
|
878 |
+
"grad_norm": 0.18496295809745789,
|
879 |
+
"learning_rate": 9.999230638894548e-05,
|
880 |
+
"loss": 2.5784,
|
881 |
+
"step": 11200
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 0.08,
|
885 |
+
"grad_norm": 0.1987435519695282,
|
886 |
+
"learning_rate": 9.99921382319577e-05,
|
887 |
+
"loss": 2.5779,
|
888 |
+
"step": 11300
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 0.08,
|
892 |
+
"grad_norm": 0.23174594342708588,
|
893 |
+
"learning_rate": 9.999196825724813e-05,
|
894 |
+
"loss": 2.5781,
|
895 |
+
"step": 11400
|
896 |
+
},
|
897 |
+
{
|
898 |
+
"epoch": 0.08,
|
899 |
+
"grad_norm": 0.18109597265720367,
|
900 |
+
"learning_rate": 9.999179646482295e-05,
|
901 |
+
"loss": 2.5785,
|
902 |
+
"step": 11500
|
903 |
+
},
|
904 |
+
{
|
905 |
+
"epoch": 0.08,
|
906 |
+
"grad_norm": 0.18909253180027008,
|
907 |
+
"learning_rate": 9.999162285468841e-05,
|
908 |
+
"loss": 2.5779,
|
909 |
+
"step": 11600
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 0.09,
|
913 |
+
"grad_norm": 0.1274472177028656,
|
914 |
+
"learning_rate": 9.999144742685083e-05,
|
915 |
+
"loss": 2.5774,
|
916 |
+
"step": 11700
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 0.09,
|
920 |
+
"grad_norm": 0.14917853474617004,
|
921 |
+
"learning_rate": 9.999127018131655e-05,
|
922 |
+
"loss": 2.5759,
|
923 |
+
"step": 11800
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 0.09,
|
927 |
+
"grad_norm": 0.13930270075798035,
|
928 |
+
"learning_rate": 9.999109111809207e-05,
|
929 |
+
"loss": 2.5769,
|
930 |
+
"step": 11900
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 0.09,
|
934 |
+
"grad_norm": 0.18673798441886902,
|
935 |
+
"learning_rate": 9.999091023718388e-05,
|
936 |
+
"loss": 2.5765,
|
937 |
+
"step": 12000
|
938 |
+
},
|
939 |
+
{
|
940 |
+
"epoch": 0.09,
|
941 |
+
"eval_loss": 2.576554775238037,
|
942 |
+
"eval_runtime": 5806.3953,
|
943 |
+
"eval_samples_per_second": 1010.102,
|
944 |
+
"eval_steps_per_second": 63.131,
|
945 |
+
"step": 12000
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 0.09,
|
949 |
+
"grad_norm": 0.23524725437164307,
|
950 |
+
"learning_rate": 9.999072753859854e-05,
|
951 |
+
"loss": 2.577,
|
952 |
+
"step": 12100
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 0.09,
|
956 |
+
"grad_norm": 0.23039047420024872,
|
957 |
+
"learning_rate": 9.99905430223427e-05,
|
958 |
+
"loss": 2.5777,
|
959 |
+
"step": 12200
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 0.09,
|
963 |
+
"grad_norm": 0.17183104157447815,
|
964 |
+
"learning_rate": 9.99903566884231e-05,
|
965 |
+
"loss": 2.5757,
|
966 |
+
"step": 12300
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 0.09,
|
970 |
+
"grad_norm": 0.22224171459674835,
|
971 |
+
"learning_rate": 9.999016853684646e-05,
|
972 |
+
"loss": 2.5775,
|
973 |
+
"step": 12400
|
974 |
+
},
|
975 |
+
{
|
976 |
+
"epoch": 0.09,
|
977 |
+
"grad_norm": 0.2552145719528198,
|
978 |
+
"learning_rate": 9.998997856761967e-05,
|
979 |
+
"loss": 2.5753,
|
980 |
+
"step": 12500
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 0.09,
|
984 |
+
"grad_norm": 0.14996840059757233,
|
985 |
+
"learning_rate": 9.998978678074961e-05,
|
986 |
+
"loss": 2.577,
|
987 |
+
"step": 12600
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 0.09,
|
991 |
+
"grad_norm": 0.22422178089618683,
|
992 |
+
"learning_rate": 9.998959317624325e-05,
|
993 |
+
"loss": 2.576,
|
994 |
+
"step": 12700
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 0.09,
|
998 |
+
"grad_norm": 0.15495595335960388,
|
999 |
+
"learning_rate": 9.998939775410767e-05,
|
1000 |
+
"loss": 2.5773,
|
1001 |
+
"step": 12800
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 0.09,
|
1005 |
+
"grad_norm": 0.21027402579784393,
|
1006 |
+
"learning_rate": 9.998920051434992e-05,
|
1007 |
+
"loss": 2.575,
|
1008 |
+
"step": 12900
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 0.09,
|
1012 |
+
"grad_norm": 0.17667286098003387,
|
1013 |
+
"learning_rate": 9.99890014569772e-05,
|
1014 |
+
"loss": 2.5765,
|
1015 |
+
"step": 13000
|
1016 |
+
},
|
1017 |
+
{
|
1018 |
+
"epoch": 0.09,
|
1019 |
+
"eval_loss": 2.574887275695801,
|
1020 |
+
"eval_runtime": 5869.3511,
|
1021 |
+
"eval_samples_per_second": 999.268,
|
1022 |
+
"eval_steps_per_second": 62.454,
|
1023 |
+
"step": 13000
|
1024 |
+
},
|
1025 |
+
{
|
1026 |
+
"epoch": 0.1,
|
1027 |
+
"grad_norm": 0.21658362448215485,
|
1028 |
+
"learning_rate": 9.998880058199675e-05,
|
1029 |
+
"loss": 2.5741,
|
1030 |
+
"step": 13100
|
1031 |
+
},
|
1032 |
+
{
|
1033 |
+
"epoch": 0.1,
|
1034 |
+
"grad_norm": 0.20807647705078125,
|
1035 |
+
"learning_rate": 9.998859788941588e-05,
|
1036 |
+
"loss": 2.5756,
|
1037 |
+
"step": 13200
|
1038 |
+
},
|
1039 |
+
{
|
1040 |
+
"epoch": 0.1,
|
1041 |
+
"grad_norm": 0.1646522879600525,
|
1042 |
+
"learning_rate": 9.998839337924195e-05,
|
1043 |
+
"loss": 2.5756,
|
1044 |
+
"step": 13300
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 0.1,
|
1048 |
+
"grad_norm": 0.166889950633049,
|
1049 |
+
"learning_rate": 9.998818705148238e-05,
|
1050 |
+
"loss": 2.5745,
|
1051 |
+
"step": 13400
|
1052 |
+
},
|
1053 |
+
{
|
1054 |
+
"epoch": 0.1,
|
1055 |
+
"grad_norm": 0.25859805941581726,
|
1056 |
+
"learning_rate": 9.998797890614469e-05,
|
1057 |
+
"loss": 2.5747,
|
1058 |
+
"step": 13500
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 0.1,
|
1062 |
+
"grad_norm": 0.20016120374202728,
|
1063 |
+
"learning_rate": 9.998776894323645e-05,
|
1064 |
+
"loss": 2.5746,
|
1065 |
+
"step": 13600
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 0.1,
|
1069 |
+
"grad_norm": 0.16718171536922455,
|
1070 |
+
"learning_rate": 9.998755716276528e-05,
|
1071 |
+
"loss": 2.574,
|
1072 |
+
"step": 13700
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 0.1,
|
1076 |
+
"grad_norm": 0.20205742120742798,
|
1077 |
+
"learning_rate": 9.998734356473892e-05,
|
1078 |
+
"loss": 2.5754,
|
1079 |
+
"step": 13800
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 0.1,
|
1083 |
+
"grad_norm": 0.14121076464653015,
|
1084 |
+
"learning_rate": 9.998712814916508e-05,
|
1085 |
+
"loss": 2.5746,
|
1086 |
+
"step": 13900
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 0.1,
|
1090 |
+
"grad_norm": 0.21463043987751007,
|
1091 |
+
"learning_rate": 9.998691091605163e-05,
|
1092 |
+
"loss": 2.5748,
|
1093 |
+
"step": 14000
|
1094 |
+
},
|
1095 |
+
{
|
1096 |
+
"epoch": 0.1,
|
1097 |
+
"eval_loss": 2.573892116546631,
|
1098 |
+
"eval_runtime": 5790.9885,
|
1099 |
+
"eval_samples_per_second": 1012.79,
|
1100 |
+
"eval_steps_per_second": 63.299,
|
1101 |
+
"step": 14000
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.1,
|
1105 |
+
"grad_norm": 0.21442490816116333,
|
1106 |
+
"learning_rate": 9.998669406490967e-05,
|
1107 |
+
"loss": 2.5752,
|
1108 |
+
"step": 14100
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.1,
|
1112 |
+
"grad_norm": 0.1974353939294815,
|
1113 |
+
"learning_rate": 9.998647321491594e-05,
|
1114 |
+
"loss": 2.5733,
|
1115 |
+
"step": 14200
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.1,
|
1119 |
+
"grad_norm": 0.17868639528751373,
|
1120 |
+
"learning_rate": 9.99862505474064e-05,
|
1121 |
+
"loss": 2.5733,
|
1122 |
+
"step": 14300
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.1,
|
1126 |
+
"grad_norm": 0.18465746939182281,
|
1127 |
+
"learning_rate": 9.998602606238913e-05,
|
1128 |
+
"loss": 2.5727,
|
1129 |
+
"step": 14400
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.11,
|
1133 |
+
"grad_norm": 0.22418785095214844,
|
1134 |
+
"learning_rate": 9.998579975987234e-05,
|
1135 |
+
"loss": 2.5739,
|
1136 |
+
"step": 14500
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.11,
|
1140 |
+
"grad_norm": 0.20102180540561676,
|
1141 |
+
"learning_rate": 9.998557163986423e-05,
|
1142 |
+
"loss": 2.5726,
|
1143 |
+
"step": 14600
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.11,
|
1147 |
+
"grad_norm": 0.1889944076538086,
|
1148 |
+
"learning_rate": 9.998534170237307e-05,
|
1149 |
+
"loss": 2.5719,
|
1150 |
+
"step": 14700
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.11,
|
1154 |
+
"grad_norm": 0.1904602348804474,
|
1155 |
+
"learning_rate": 9.998510994740727e-05,
|
1156 |
+
"loss": 2.5719,
|
1157 |
+
"step": 14800
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.11,
|
1161 |
+
"grad_norm": 0.20775534212589264,
|
1162 |
+
"learning_rate": 9.998487637497522e-05,
|
1163 |
+
"loss": 2.5725,
|
1164 |
+
"step": 14900
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.11,
|
1168 |
+
"grad_norm": 0.203061044216156,
|
1169 |
+
"learning_rate": 9.998464334798072e-05,
|
1170 |
+
"loss": 2.5733,
|
1171 |
+
"step": 15000
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.11,
|
1175 |
+
"eval_loss": 2.5725255012512207,
|
1176 |
+
"eval_runtime": 5867.3015,
|
1177 |
+
"eval_samples_per_second": 999.617,
|
1178 |
+
"eval_steps_per_second": 62.476,
|
1179 |
+
"step": 15000
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 0.11,
|
1183 |
+
"grad_norm": 0.1729467511177063,
|
1184 |
+
"learning_rate": 9.998440615881618e-05,
|
1185 |
+
"loss": 2.5729,
|
1186 |
+
"step": 15100
|
1187 |
+
},
|
1188 |
+
{
|
1189 |
+
"epoch": 0.11,
|
1190 |
+
"grad_norm": 0.1721724420785904,
|
1191 |
+
"learning_rate": 9.998416715221101e-05,
|
1192 |
+
"loss": 2.5719,
|
1193 |
+
"step": 15200
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 0.11,
|
1197 |
+
"grad_norm": 0.170950248837471,
|
1198 |
+
"learning_rate": 9.998392632817387e-05,
|
1199 |
+
"loss": 2.5725,
|
1200 |
+
"step": 15300
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 0.11,
|
1204 |
+
"grad_norm": 0.1992483288049698,
|
1205 |
+
"learning_rate": 9.998368368671353e-05,
|
1206 |
+
"loss": 2.573,
|
1207 |
+
"step": 15400
|
1208 |
+
},
|
1209 |
+
{
|
1210 |
+
"epoch": 0.11,
|
1211 |
+
"grad_norm": 0.19748467206954956,
|
1212 |
+
"learning_rate": 9.998343922783881e-05,
|
1213 |
+
"loss": 2.5719,
|
1214 |
+
"step": 15500
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 0.11,
|
1218 |
+
"grad_norm": 0.22308781743049622,
|
1219 |
+
"learning_rate": 9.99831929515586e-05,
|
1220 |
+
"loss": 2.5712,
|
1221 |
+
"step": 15600
|
1222 |
+
},
|
1223 |
+
{
|
1224 |
+
"epoch": 0.11,
|
1225 |
+
"grad_norm": 0.17296306788921356,
|
1226 |
+
"learning_rate": 9.998294485788187e-05,
|
1227 |
+
"loss": 2.5709,
|
1228 |
+
"step": 15700
|
1229 |
+
},
|
1230 |
+
{
|
1231 |
+
"epoch": 0.11,
|
1232 |
+
"grad_norm": 0.2266322374343872,
|
1233 |
+
"learning_rate": 9.99826949468176e-05,
|
1234 |
+
"loss": 2.5716,
|
1235 |
+
"step": 15800
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 0.12,
|
1239 |
+
"grad_norm": 0.1890050321817398,
|
1240 |
+
"learning_rate": 9.998244321837492e-05,
|
1241 |
+
"loss": 2.5714,
|
1242 |
+
"step": 15900
|
1243 |
+
},
|
1244 |
+
{
|
1245 |
+
"epoch": 0.12,
|
1246 |
+
"grad_norm": 0.14952068030834198,
|
1247 |
+
"learning_rate": 9.998218967256294e-05,
|
1248 |
+
"loss": 2.5724,
|
1249 |
+
"step": 16000
|
1250 |
+
},
|
1251 |
+
{
|
1252 |
+
"epoch": 0.12,
|
1253 |
+
"eval_loss": 2.570789098739624,
|
1254 |
+
"eval_runtime": 5730.232,
|
1255 |
+
"eval_samples_per_second": 1023.528,
|
1256 |
+
"eval_steps_per_second": 63.971,
|
1257 |
+
"step": 16000
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 0.12,
|
1261 |
+
"grad_norm": 0.20020237565040588,
|
1262 |
+
"learning_rate": 9.998193430939093e-05,
|
1263 |
+
"loss": 2.5712,
|
1264 |
+
"step": 16100
|
1265 |
+
},
|
1266 |
+
{
|
1267 |
+
"epoch": 0.12,
|
1268 |
+
"grad_norm": 0.2040109634399414,
|
1269 |
+
"learning_rate": 9.998167712886813e-05,
|
1270 |
+
"loss": 2.5722,
|
1271 |
+
"step": 16200
|
1272 |
+
},
|
1273 |
+
{
|
1274 |
+
"epoch": 0.12,
|
1275 |
+
"grad_norm": 0.14661578834056854,
|
1276 |
+
"learning_rate": 9.998141813100392e-05,
|
1277 |
+
"loss": 2.5713,
|
1278 |
+
"step": 16300
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 0.12,
|
1282 |
+
"grad_norm": 0.17079943418502808,
|
1283 |
+
"learning_rate": 9.998115731580771e-05,
|
1284 |
+
"loss": 2.5718,
|
1285 |
+
"step": 16400
|
1286 |
+
},
|
1287 |
+
{
|
1288 |
+
"epoch": 0.12,
|
1289 |
+
"grad_norm": 0.17369835078716278,
|
1290 |
+
"learning_rate": 9.998089468328898e-05,
|
1291 |
+
"loss": 2.5721,
|
1292 |
+
"step": 16500
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 0.12,
|
1296 |
+
"grad_norm": 0.16338428854942322,
|
1297 |
+
"learning_rate": 9.998063023345725e-05,
|
1298 |
+
"loss": 2.5718,
|
1299 |
+
"step": 16600
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 0.12,
|
1303 |
+
"grad_norm": 0.16794905066490173,
|
1304 |
+
"learning_rate": 9.99803639663222e-05,
|
1305 |
+
"loss": 2.5708,
|
1306 |
+
"step": 16700
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 0.12,
|
1310 |
+
"grad_norm": 0.20138411223888397,
|
1311 |
+
"learning_rate": 9.998009588189345e-05,
|
1312 |
+
"loss": 2.5696,
|
1313 |
+
"step": 16800
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 0.12,
|
1317 |
+
"grad_norm": 0.23276585340499878,
|
1318 |
+
"learning_rate": 9.99798259801808e-05,
|
1319 |
+
"loss": 2.5691,
|
1320 |
+
"step": 16900
|
1321 |
+
},
|
1322 |
+
{
|
1323 |
+
"epoch": 0.12,
|
1324 |
+
"grad_norm": 0.1607973575592041,
|
1325 |
+
"learning_rate": 9.997955426119402e-05,
|
1326 |
+
"loss": 2.5714,
|
1327 |
+
"step": 17000
|
1328 |
+
},
|
1329 |
+
{
|
1330 |
+
"epoch": 0.12,
|
1331 |
+
"eval_loss": 2.569423198699951,
|
1332 |
+
"eval_runtime": 5840.5846,
|
1333 |
+
"eval_samples_per_second": 1004.189,
|
1334 |
+
"eval_steps_per_second": 62.762,
|
1335 |
+
"step": 17000
|
1336 |
+
},
|
1337 |
+
{
|
1338 |
+
"epoch": 0.12,
|
1339 |
+
"grad_norm": 0.1984170377254486,
|
1340 |
+
"learning_rate": 9.997928072494302e-05,
|
1341 |
+
"loss": 2.5703,
|
1342 |
+
"step": 17100
|
1343 |
+
},
|
1344 |
+
{
|
1345 |
+
"epoch": 0.13,
|
1346 |
+
"grad_norm": 0.22356334328651428,
|
1347 |
+
"learning_rate": 9.997900537143772e-05,
|
1348 |
+
"loss": 2.5711,
|
1349 |
+
"step": 17200
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 0.13,
|
1353 |
+
"grad_norm": 0.1815560907125473,
|
1354 |
+
"learning_rate": 9.997873098139096e-05,
|
1355 |
+
"loss": 2.5697,
|
1356 |
+
"step": 17300
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 0.13,
|
1360 |
+
"grad_norm": 0.27280667424201965,
|
1361 |
+
"learning_rate": 9.997845201157949e-05,
|
1362 |
+
"loss": 2.5697,
|
1363 |
+
"step": 17400
|
1364 |
+
},
|
1365 |
+
{
|
1366 |
+
"epoch": 0.13,
|
1367 |
+
"grad_norm": 0.15276475250720978,
|
1368 |
+
"learning_rate": 9.997817122454387e-05,
|
1369 |
+
"loss": 2.5675,
|
1370 |
+
"step": 17500
|
1371 |
+
},
|
1372 |
+
{
|
1373 |
+
"epoch": 0.13,
|
1374 |
+
"grad_norm": 0.19267494976520538,
|
1375 |
+
"learning_rate": 9.997788862029429e-05,
|
1376 |
+
"loss": 2.569,
|
1377 |
+
"step": 17600
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 0.13,
|
1381 |
+
"grad_norm": 0.1580013483762741,
|
1382 |
+
"learning_rate": 9.997760419884105e-05,
|
1383 |
+
"loss": 2.5704,
|
1384 |
+
"step": 17700
|
1385 |
+
},
|
1386 |
+
{
|
1387 |
+
"epoch": 0.13,
|
1388 |
+
"grad_norm": 0.21980580687522888,
|
1389 |
+
"learning_rate": 9.997731796019448e-05,
|
1390 |
+
"loss": 2.5702,
|
1391 |
+
"step": 17800
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 0.13,
|
1395 |
+
"grad_norm": 0.2392682433128357,
|
1396 |
+
"learning_rate": 9.997702990436498e-05,
|
1397 |
+
"loss": 2.5696,
|
1398 |
+
"step": 17900
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 0.13,
|
1402 |
+
"grad_norm": 0.2011999785900116,
|
1403 |
+
"learning_rate": 9.997674003136303e-05,
|
1404 |
+
"loss": 2.5682,
|
1405 |
+
"step": 18000
|
1406 |
+
},
|
1407 |
+
{
|
1408 |
+
"epoch": 0.13,
|
1409 |
+
"eval_loss": 2.5681583881378174,
|
1410 |
+
"eval_runtime": 5608.808,
|
1411 |
+
"eval_samples_per_second": 1045.686,
|
1412 |
+
"eval_steps_per_second": 65.355,
|
1413 |
+
"step": 18000
|
1414 |
+
},
|
1415 |
+
{
|
1416 |
+
"epoch": 0.13,
|
1417 |
+
"grad_norm": 0.23138105869293213,
|
1418 |
+
"learning_rate": 9.997644834119919e-05,
|
1419 |
+
"loss": 2.5685,
|
1420 |
+
"step": 18100
|
1421 |
+
},
|
1422 |
+
{
|
1423 |
+
"epoch": 0.13,
|
1424 |
+
"grad_norm": 0.19521869719028473,
|
1425 |
+
"learning_rate": 9.997615483388406e-05,
|
1426 |
+
"loss": 2.5685,
|
1427 |
+
"step": 18200
|
1428 |
+
},
|
1429 |
+
{
|
1430 |
+
"epoch": 0.13,
|
1431 |
+
"grad_norm": 0.22132566571235657,
|
1432 |
+
"learning_rate": 9.997585950942827e-05,
|
1433 |
+
"loss": 2.5687,
|
1434 |
+
"step": 18300
|
1435 |
+
},
|
1436 |
+
{
|
1437 |
+
"epoch": 0.13,
|
1438 |
+
"grad_norm": 0.19450217485427856,
|
1439 |
+
"learning_rate": 9.99755623678426e-05,
|
1440 |
+
"loss": 2.5666,
|
1441 |
+
"step": 18400
|
1442 |
+
},
|
1443 |
+
{
|
1444 |
+
"epoch": 0.13,
|
1445 |
+
"grad_norm": 0.20810718834400177,
|
1446 |
+
"learning_rate": 9.997526340913785e-05,
|
1447 |
+
"loss": 2.5681,
|
1448 |
+
"step": 18500
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 0.14,
|
1452 |
+
"grad_norm": 0.1711287647485733,
|
1453 |
+
"learning_rate": 9.997496263332487e-05,
|
1454 |
+
"loss": 2.5666,
|
1455 |
+
"step": 18600
|
1456 |
+
},
|
1457 |
+
{
|
1458 |
+
"epoch": 0.14,
|
1459 |
+
"grad_norm": 0.15737326443195343,
|
1460 |
+
"learning_rate": 9.997466004041462e-05,
|
1461 |
+
"loss": 2.5663,
|
1462 |
+
"step": 18700
|
1463 |
+
},
|
1464 |
+
{
|
1465 |
+
"epoch": 0.14,
|
1466 |
+
"grad_norm": 0.19085867702960968,
|
1467 |
+
"learning_rate": 9.997435563041809e-05,
|
1468 |
+
"loss": 2.569,
|
1469 |
+
"step": 18800
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 0.14,
|
1473 |
+
"grad_norm": 0.20687063038349152,
|
1474 |
+
"learning_rate": 9.997404940334637e-05,
|
1475 |
+
"loss": 2.5654,
|
1476 |
+
"step": 18900
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 0.14,
|
1480 |
+
"grad_norm": 0.21126438677310944,
|
1481 |
+
"learning_rate": 9.997374135921054e-05,
|
1482 |
+
"loss": 2.5677,
|
1483 |
+
"step": 19000
|
1484 |
+
},
|
1485 |
+
{
|
1486 |
+
"epoch": 0.14,
|
1487 |
+
"eval_loss": 2.5664379596710205,
|
1488 |
+
"eval_runtime": 5437.5739,
|
1489 |
+
"eval_samples_per_second": 1078.616,
|
1490 |
+
"eval_steps_per_second": 67.414,
|
1491 |
+
"step": 19000
|
1492 |
+
},
|
1493 |
+
{
|
1494 |
+
"epoch": 0.14,
|
1495 |
+
"grad_norm": 0.1988440304994583,
|
1496 |
+
"learning_rate": 9.997343149802186e-05,
|
1497 |
+
"loss": 2.5667,
|
1498 |
+
"step": 19100
|
1499 |
+
},
|
1500 |
+
{
|
1501 |
+
"epoch": 0.14,
|
1502 |
+
"grad_norm": 0.2025243192911148,
|
1503 |
+
"learning_rate": 9.997311981979155e-05,
|
1504 |
+
"loss": 2.5674,
|
1505 |
+
"step": 19200
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 0.14,
|
1509 |
+
"grad_norm": 0.15592768788337708,
|
1510 |
+
"learning_rate": 9.997280632453097e-05,
|
1511 |
+
"loss": 2.5679,
|
1512 |
+
"step": 19300
|
1513 |
+
},
|
1514 |
+
{
|
1515 |
+
"epoch": 0.14,
|
1516 |
+
"grad_norm": 0.18881608545780182,
|
1517 |
+
"learning_rate": 9.997249101225153e-05,
|
1518 |
+
"loss": 2.5664,
|
1519 |
+
"step": 19400
|
1520 |
+
},
|
1521 |
+
{
|
1522 |
+
"epoch": 0.14,
|
1523 |
+
"grad_norm": 0.2082224190235138,
|
1524 |
+
"learning_rate": 9.997217706325169e-05,
|
1525 |
+
"loss": 2.5653,
|
1526 |
+
"step": 19500
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 0.14,
|
1530 |
+
"grad_norm": 0.20503577589988708,
|
1531 |
+
"learning_rate": 9.997185813513884e-05,
|
1532 |
+
"loss": 2.5675,
|
1533 |
+
"step": 19600
|
1534 |
+
},
|
1535 |
+
{
|
1536 |
+
"epoch": 0.14,
|
1537 |
+
"grad_norm": 0.157631978392601,
|
1538 |
+
"learning_rate": 9.997153739004159e-05,
|
1539 |
+
"loss": 2.5655,
|
1540 |
+
"step": 19700
|
1541 |
+
},
|
1542 |
+
{
|
1543 |
+
"epoch": 0.14,
|
1544 |
+
"grad_norm": 0.1750982105731964,
|
1545 |
+
"learning_rate": 9.997121482797162e-05,
|
1546 |
+
"loss": 2.566,
|
1547 |
+
"step": 19800
|
1548 |
+
},
|
1549 |
+
{
|
1550 |
+
"epoch": 0.14,
|
1551 |
+
"grad_norm": 0.2016943246126175,
|
1552 |
+
"learning_rate": 9.997089044894064e-05,
|
1553 |
+
"loss": 2.564,
|
1554 |
+
"step": 19900
|
1555 |
+
},
|
1556 |
+
{
|
1557 |
+
"epoch": 0.15,
|
1558 |
+
"grad_norm": 0.23606260120868683,
|
1559 |
+
"learning_rate": 9.997056425296043e-05,
|
1560 |
+
"loss": 2.5643,
|
1561 |
+
"step": 20000
|
1562 |
+
},
|
1563 |
+
{
|
1564 |
+
"epoch": 0.15,
|
1565 |
+
"eval_loss": 2.564971446990967,
|
1566 |
+
"eval_runtime": 5311.9496,
|
1567 |
+
"eval_samples_per_second": 1104.124,
|
1568 |
+
"eval_steps_per_second": 69.008,
|
1569 |
+
"step": 20000
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 0.15,
|
1573 |
+
"grad_norm": 0.21572504937648773,
|
1574 |
+
"learning_rate": 9.997023624004287e-05,
|
1575 |
+
"loss": 2.5655,
|
1576 |
+
"step": 20100
|
1577 |
+
},
|
1578 |
+
{
|
1579 |
+
"epoch": 0.15,
|
1580 |
+
"grad_norm": 0.1901276409626007,
|
1581 |
+
"learning_rate": 9.996990641019987e-05,
|
1582 |
+
"loss": 2.5646,
|
1583 |
+
"step": 20200
|
1584 |
+
},
|
1585 |
+
{
|
1586 |
+
"epoch": 0.15,
|
1587 |
+
"grad_norm": 0.1661011129617691,
|
1588 |
+
"learning_rate": 9.996957476344345e-05,
|
1589 |
+
"loss": 2.5656,
|
1590 |
+
"step": 20300
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 0.15,
|
1594 |
+
"grad_norm": 0.1857927441596985,
|
1595 |
+
"learning_rate": 9.996924129978566e-05,
|
1596 |
+
"loss": 2.5641,
|
1597 |
+
"step": 20400
|
1598 |
+
},
|
1599 |
+
{
|
1600 |
+
"epoch": 0.15,
|
1601 |
+
"grad_norm": 0.17554591596126556,
|
1602 |
+
"learning_rate": 9.996890601923861e-05,
|
1603 |
+
"loss": 2.565,
|
1604 |
+
"step": 20500
|
1605 |
+
},
|
1606 |
+
{
|
1607 |
+
"epoch": 0.15,
|
1608 |
+
"grad_norm": 0.19168546795845032,
|
1609 |
+
"learning_rate": 9.99685689218145e-05,
|
1610 |
+
"loss": 2.5634,
|
1611 |
+
"step": 20600
|
1612 |
+
},
|
1613 |
+
{
|
1614 |
+
"epoch": 0.15,
|
1615 |
+
"grad_norm": 0.24985399842262268,
|
1616 |
+
"learning_rate": 9.996823000752557e-05,
|
1617 |
+
"loss": 2.5645,
|
1618 |
+
"step": 20700
|
1619 |
+
},
|
1620 |
+
{
|
1621 |
+
"epoch": 0.15,
|
1622 |
+
"grad_norm": 0.17692847549915314,
|
1623 |
+
"learning_rate": 9.996788927638418e-05,
|
1624 |
+
"loss": 2.5623,
|
1625 |
+
"step": 20800
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 0.15,
|
1629 |
+
"grad_norm": 0.19862769544124603,
|
1630 |
+
"learning_rate": 9.996754672840269e-05,
|
1631 |
+
"loss": 2.5637,
|
1632 |
+
"step": 20900
|
1633 |
+
},
|
1634 |
+
{
|
1635 |
+
"epoch": 0.15,
|
1636 |
+
"grad_norm": 0.15231578052043915,
|
1637 |
+
"learning_rate": 9.996720236359356e-05,
|
1638 |
+
"loss": 2.5629,
|
1639 |
+
"step": 21000
|
1640 |
+
},
|
1641 |
+
{
|
1642 |
+
"epoch": 0.15,
|
1643 |
+
"eval_loss": 2.5632758140563965,
|
1644 |
+
"eval_runtime": 5223.2614,
|
1645 |
+
"eval_samples_per_second": 1122.872,
|
1646 |
+
"eval_steps_per_second": 70.18,
|
1647 |
+
"step": 21000
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.15,
|
1651 |
+
"grad_norm": 0.15703310072422028,
|
1652 |
+
"learning_rate": 9.996685618196933e-05,
|
1653 |
+
"loss": 2.5647,
|
1654 |
+
"step": 21100
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.15,
|
1658 |
+
"grad_norm": 0.21033476293087006,
|
1659 |
+
"learning_rate": 9.996650818354254e-05,
|
1660 |
+
"loss": 2.5616,
|
1661 |
+
"step": 21200
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.15,
|
1665 |
+
"grad_norm": 0.17880433797836304,
|
1666 |
+
"learning_rate": 9.99661583683259e-05,
|
1667 |
+
"loss": 2.5632,
|
1668 |
+
"step": 21300
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.16,
|
1672 |
+
"grad_norm": 0.17895223200321198,
|
1673 |
+
"learning_rate": 9.996580673633208e-05,
|
1674 |
+
"loss": 2.5612,
|
1675 |
+
"step": 21400
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.16,
|
1679 |
+
"grad_norm": 0.1816408783197403,
|
1680 |
+
"learning_rate": 9.996545683105445e-05,
|
1681 |
+
"loss": 2.5627,
|
1682 |
+
"step": 21500
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.16,
|
1686 |
+
"grad_norm": 0.17461837828159332,
|
1687 |
+
"learning_rate": 9.996510158371221e-05,
|
1688 |
+
"loss": 2.5622,
|
1689 |
+
"step": 21600
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.16,
|
1693 |
+
"grad_norm": 0.22741451859474182,
|
1694 |
+
"learning_rate": 9.996474451963123e-05,
|
1695 |
+
"loss": 2.5626,
|
1696 |
+
"step": 21700
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.16,
|
1700 |
+
"grad_norm": 0.2556922733783722,
|
1701 |
+
"learning_rate": 9.996438563882451e-05,
|
1702 |
+
"loss": 2.5627,
|
1703 |
+
"step": 21800
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.16,
|
1707 |
+
"grad_norm": 0.18055401742458344,
|
1708 |
+
"learning_rate": 9.99640249413051e-05,
|
1709 |
+
"loss": 2.5623,
|
1710 |
+
"step": 21900
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.16,
|
1714 |
+
"grad_norm": 0.1370435506105423,
|
1715 |
+
"learning_rate": 9.996366242708609e-05,
|
1716 |
+
"loss": 2.5629,
|
1717 |
+
"step": 22000
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.16,
|
1721 |
+
"eval_loss": 2.5613725185394287,
|
1722 |
+
"eval_runtime": 5233.9822,
|
1723 |
+
"eval_samples_per_second": 1120.572,
|
1724 |
+
"eval_steps_per_second": 70.036,
|
1725 |
+
"step": 22000
|
1726 |
+
},
|
1727 |
+
{
|
1728 |
+
"epoch": 0.16,
|
1729 |
+
"grad_norm": 0.150621235370636,
|
1730 |
+
"learning_rate": 9.996329809618068e-05,
|
1731 |
+
"loss": 2.5595,
|
1732 |
+
"step": 22100
|
1733 |
+
},
|
1734 |
+
{
|
1735 |
+
"epoch": 0.16,
|
1736 |
+
"grad_norm": 0.23477642238140106,
|
1737 |
+
"learning_rate": 9.996293194860211e-05,
|
1738 |
+
"loss": 2.561,
|
1739 |
+
"step": 22200
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 0.16,
|
1743 |
+
"grad_norm": 0.20428013801574707,
|
1744 |
+
"learning_rate": 9.996256398436372e-05,
|
1745 |
+
"loss": 2.5589,
|
1746 |
+
"step": 22300
|
1747 |
+
},
|
1748 |
+
{
|
1749 |
+
"epoch": 0.16,
|
1750 |
+
"grad_norm": 0.18161137402057648,
|
1751 |
+
"learning_rate": 9.996219420347885e-05,
|
1752 |
+
"loss": 2.5618,
|
1753 |
+
"step": 22400
|
1754 |
+
},
|
1755 |
+
{
|
1756 |
+
"epoch": 0.16,
|
1757 |
+
"grad_norm": 0.18925230205059052,
|
1758 |
+
"learning_rate": 9.996182260596096e-05,
|
1759 |
+
"loss": 2.5579,
|
1760 |
+
"step": 22500
|
1761 |
+
},
|
1762 |
+
{
|
1763 |
+
"epoch": 0.16,
|
1764 |
+
"grad_norm": 0.20608845353126526,
|
1765 |
+
"learning_rate": 9.996144919182355e-05,
|
1766 |
+
"loss": 2.5592,
|
1767 |
+
"step": 22600
|
1768 |
+
},
|
1769 |
+
{
|
1770 |
+
"epoch": 0.17,
|
1771 |
+
"grad_norm": 0.19250337779521942,
|
1772 |
+
"learning_rate": 9.996107396108024e-05,
|
1773 |
+
"loss": 2.5596,
|
1774 |
+
"step": 22700
|
1775 |
+
},
|
1776 |
+
{
|
1777 |
+
"epoch": 0.17,
|
1778 |
+
"grad_norm": 0.2164393663406372,
|
1779 |
+
"learning_rate": 9.996069691374462e-05,
|
1780 |
+
"loss": 2.5596,
|
1781 |
+
"step": 22800
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 0.17,
|
1785 |
+
"grad_norm": 0.16476331651210785,
|
1786 |
+
"learning_rate": 9.996031804983043e-05,
|
1787 |
+
"loss": 2.5606,
|
1788 |
+
"step": 22900
|
1789 |
+
},
|
1790 |
+
{
|
1791 |
+
"epoch": 0.17,
|
1792 |
+
"grad_norm": 0.15099835395812988,
|
1793 |
+
"learning_rate": 9.995993736935145e-05,
|
1794 |
+
"loss": 2.5612,
|
1795 |
+
"step": 23000
|
1796 |
+
},
|
1797 |
+
{
|
1798 |
+
"epoch": 0.17,
|
1799 |
+
"eval_loss": 2.5598702430725098,
|
1800 |
+
"eval_runtime": 6082.4008,
|
1801 |
+
"eval_samples_per_second": 964.266,
|
1802 |
+
"eval_steps_per_second": 60.267,
|
1803 |
+
"step": 23000
|
1804 |
+
},
|
1805 |
+
{
|
1806 |
+
"epoch": 0.17,
|
1807 |
+
"grad_norm": 0.1863376796245575,
|
1808 |
+
"learning_rate": 9.99595548723215e-05,
|
1809 |
+
"loss": 2.5593,
|
1810 |
+
"step": 23100
|
1811 |
+
},
|
1812 |
+
{
|
1813 |
+
"epoch": 0.17,
|
1814 |
+
"grad_norm": 0.19632737338542938,
|
1815 |
+
"learning_rate": 9.99591705587545e-05,
|
1816 |
+
"loss": 2.5594,
|
1817 |
+
"step": 23200
|
1818 |
+
},
|
1819 |
+
{
|
1820 |
+
"epoch": 0.17,
|
1821 |
+
"grad_norm": 0.18330197036266327,
|
1822 |
+
"learning_rate": 9.995878442866442e-05,
|
1823 |
+
"loss": 2.5598,
|
1824 |
+
"step": 23300
|
1825 |
+
},
|
1826 |
+
{
|
1827 |
+
"epoch": 0.17,
|
1828 |
+
"grad_norm": 0.23508641123771667,
|
1829 |
+
"learning_rate": 9.995839648206531e-05,
|
1830 |
+
"loss": 2.5579,
|
1831 |
+
"step": 23400
|
1832 |
+
},
|
1833 |
+
{
|
1834 |
+
"epoch": 0.17,
|
1835 |
+
"grad_norm": 0.2019587904214859,
|
1836 |
+
"learning_rate": 9.99580106255938e-05,
|
1837 |
+
"loss": 2.5584,
|
1838 |
+
"step": 23500
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 0.17,
|
1842 |
+
"grad_norm": 0.20707851648330688,
|
1843 |
+
"learning_rate": 9.995761906418372e-05,
|
1844 |
+
"loss": 2.5589,
|
1845 |
+
"step": 23600
|
1846 |
+
},
|
1847 |
+
{
|
1848 |
+
"epoch": 0.17,
|
1849 |
+
"grad_norm": 0.16306054592132568,
|
1850 |
+
"learning_rate": 9.995722568630698e-05,
|
1851 |
+
"loss": 2.5576,
|
1852 |
+
"step": 23700
|
1853 |
+
},
|
1854 |
+
{
|
1855 |
+
"epoch": 0.17,
|
1856 |
+
"grad_norm": 0.21417422592639923,
|
1857 |
+
"learning_rate": 9.995683049197788e-05,
|
1858 |
+
"loss": 2.5574,
|
1859 |
+
"step": 23800
|
1860 |
+
},
|
1861 |
+
{
|
1862 |
+
"epoch": 0.17,
|
1863 |
+
"grad_norm": 0.21078960597515106,
|
1864 |
+
"learning_rate": 9.99564334812108e-05,
|
1865 |
+
"loss": 2.5588,
|
1866 |
+
"step": 23900
|
1867 |
+
},
|
1868 |
+
{
|
1869 |
+
"epoch": 0.17,
|
1870 |
+
"grad_norm": 0.17299005389213562,
|
1871 |
+
"learning_rate": 9.995603465402014e-05,
|
1872 |
+
"loss": 2.5567,
|
1873 |
+
"step": 24000
|
1874 |
+
},
|
1875 |
+
{
|
1876 |
+
"epoch": 0.17,
|
1877 |
+
"eval_loss": 2.5584349632263184,
|
1878 |
+
"eval_runtime": 5391.8597,
|
1879 |
+
"eval_samples_per_second": 1087.761,
|
1880 |
+
"eval_steps_per_second": 67.985,
|
1881 |
+
"step": 24000
|
1882 |
+
},
|
1883 |
+
{
|
1884 |
+
"epoch": 0.18,
|
1885 |
+
"grad_norm": 0.1550675630569458,
|
1886 |
+
"learning_rate": 9.995563401042043e-05,
|
1887 |
+
"loss": 2.5576,
|
1888 |
+
"step": 24100
|
1889 |
+
},
|
1890 |
+
{
|
1891 |
+
"epoch": 0.18,
|
1892 |
+
"grad_norm": 0.21063561737537384,
|
1893 |
+
"learning_rate": 9.995523155042623e-05,
|
1894 |
+
"loss": 2.5585,
|
1895 |
+
"step": 24200
|
1896 |
+
},
|
1897 |
+
{
|
1898 |
+
"epoch": 0.18,
|
1899 |
+
"grad_norm": 0.19711926579475403,
|
1900 |
+
"learning_rate": 9.995482727405219e-05,
|
1901 |
+
"loss": 2.5589,
|
1902 |
+
"step": 24300
|
1903 |
+
},
|
1904 |
+
{
|
1905 |
+
"epoch": 0.18,
|
1906 |
+
"grad_norm": 0.19466377794742584,
|
1907 |
+
"learning_rate": 9.995442118131297e-05,
|
1908 |
+
"loss": 2.5579,
|
1909 |
+
"step": 24400
|
1910 |
+
},
|
1911 |
+
{
|
1912 |
+
"epoch": 0.18,
|
1913 |
+
"grad_norm": 0.1623954474925995,
|
1914 |
+
"learning_rate": 9.995401327222338e-05,
|
1915 |
+
"loss": 2.5577,
|
1916 |
+
"step": 24500
|
1917 |
+
},
|
1918 |
+
{
|
1919 |
+
"epoch": 0.18,
|
1920 |
+
"grad_norm": 0.18352490663528442,
|
1921 |
+
"learning_rate": 9.995360354679822e-05,
|
1922 |
+
"loss": 2.5583,
|
1923 |
+
"step": 24600
|
1924 |
+
},
|
1925 |
+
{
|
1926 |
+
"epoch": 0.18,
|
1927 |
+
"grad_norm": 0.17624753713607788,
|
1928 |
+
"learning_rate": 9.99531920050524e-05,
|
1929 |
+
"loss": 2.5577,
|
1930 |
+
"step": 24700
|
1931 |
+
},
|
1932 |
+
{
|
1933 |
+
"epoch": 0.18,
|
1934 |
+
"grad_norm": 0.15336942672729492,
|
1935 |
+
"learning_rate": 9.995277864700089e-05,
|
1936 |
+
"loss": 2.5578,
|
1937 |
+
"step": 24800
|
1938 |
+
},
|
1939 |
+
{
|
1940 |
+
"epoch": 0.18,
|
1941 |
+
"grad_norm": 0.17434526979923248,
|
1942 |
+
"learning_rate": 9.99523634726587e-05,
|
1943 |
+
"loss": 2.5569,
|
1944 |
+
"step": 24900
|
1945 |
+
},
|
1946 |
+
{
|
1947 |
+
"epoch": 0.18,
|
1948 |
+
"grad_norm": 0.21592725813388824,
|
1949 |
+
"learning_rate": 9.995194648204095e-05,
|
1950 |
+
"loss": 2.5543,
|
1951 |
+
"step": 25000
|
1952 |
+
},
|
1953 |
+
{
|
1954 |
+
"epoch": 0.18,
|
1955 |
+
"eval_loss": 2.5564043521881104,
|
1956 |
+
"eval_runtime": 5516.1995,
|
1957 |
+
"eval_samples_per_second": 1063.242,
|
1958 |
+
"eval_steps_per_second": 66.453,
|
1959 |
+
"step": 25000
|
1960 |
+
},
|
1961 |
+
{
|
1962 |
+
"epoch": 0.18,
|
1963 |
+
"grad_norm": 0.15827420353889465,
|
1964 |
+
"learning_rate": 9.99515276751628e-05,
|
1965 |
+
"loss": 2.5554,
|
1966 |
+
"step": 25100
|
1967 |
+
},
|
1968 |
+
{
|
1969 |
+
"epoch": 0.18,
|
1970 |
+
"grad_norm": 0.17489619553089142,
|
1971 |
+
"learning_rate": 9.995110705203945e-05,
|
1972 |
+
"loss": 2.5566,
|
1973 |
+
"step": 25200
|
1974 |
+
},
|
1975 |
+
{
|
1976 |
+
"epoch": 0.18,
|
1977 |
+
"grad_norm": 0.18893638253211975,
|
1978 |
+
"learning_rate": 9.995068461268622e-05,
|
1979 |
+
"loss": 2.5561,
|
1980 |
+
"step": 25300
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 0.18,
|
1984 |
+
"grad_norm": 0.19673244655132294,
|
1985 |
+
"learning_rate": 9.995026035711845e-05,
|
1986 |
+
"loss": 2.5569,
|
1987 |
+
"step": 25400
|
1988 |
+
},
|
1989 |
+
{
|
1990 |
+
"epoch": 0.19,
|
1991 |
+
"grad_norm": 0.18465718626976013,
|
1992 |
+
"learning_rate": 9.994983855505939e-05,
|
1993 |
+
"loss": 2.5554,
|
1994 |
+
"step": 25500
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 0.19,
|
1998 |
+
"grad_norm": 0.1976306438446045,
|
1999 |
+
"learning_rate": 9.994941068527068e-05,
|
2000 |
+
"loss": 2.5561,
|
2001 |
+
"step": 25600
|
2002 |
+
},
|
2003 |
+
{
|
2004 |
+
"epoch": 0.19,
|
2005 |
+
"grad_norm": 0.1467328816652298,
|
2006 |
+
"learning_rate": 9.994898099931376e-05,
|
2007 |
+
"loss": 2.5542,
|
2008 |
+
"step": 25700
|
2009 |
+
},
|
2010 |
+
{
|
2011 |
+
"epoch": 0.19,
|
2012 |
+
"grad_norm": 0.20342576503753662,
|
2013 |
+
"learning_rate": 9.994854949720426e-05,
|
2014 |
+
"loss": 2.555,
|
2015 |
+
"step": 25800
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 0.19,
|
2019 |
+
"grad_norm": 0.20438461005687714,
|
2020 |
+
"learning_rate": 9.994811617895786e-05,
|
2021 |
+
"loss": 2.5553,
|
2022 |
+
"step": 25900
|
2023 |
+
},
|
2024 |
+
{
|
2025 |
+
"epoch": 0.19,
|
2026 |
+
"grad_norm": 0.16016176342964172,
|
2027 |
+
"learning_rate": 9.994768104459032e-05,
|
2028 |
+
"loss": 2.5568,
|
2029 |
+
"step": 26000
|
2030 |
+
},
|
2031 |
+
{
|
2032 |
+
"epoch": 0.19,
|
2033 |
+
"eval_loss": 2.554506778717041,
|
2034 |
+
"eval_runtime": 5432.0955,
|
2035 |
+
"eval_samples_per_second": 1079.704,
|
2036 |
+
"eval_steps_per_second": 67.482,
|
2037 |
+
"step": 26000
|
2038 |
+
},
|
2039 |
+
{
|
2040 |
+
"epoch": 0.19,
|
2041 |
+
"grad_norm": 0.14238539338111877,
|
2042 |
+
"learning_rate": 9.994724409411746e-05,
|
2043 |
+
"loss": 2.5552,
|
2044 |
+
"step": 26100
|
2045 |
+
},
|
2046 |
+
{
|
2047 |
+
"epoch": 0.19,
|
2048 |
+
"grad_norm": 0.23573274910449982,
|
2049 |
+
"learning_rate": 9.994680532755518e-05,
|
2050 |
+
"loss": 2.5523,
|
2051 |
+
"step": 26200
|
2052 |
+
},
|
2053 |
+
{
|
2054 |
+
"epoch": 0.19,
|
2055 |
+
"grad_norm": 0.19863677024841309,
|
2056 |
+
"learning_rate": 9.994636474491942e-05,
|
2057 |
+
"loss": 2.5529,
|
2058 |
+
"step": 26300
|
2059 |
+
},
|
2060 |
+
{
|
2061 |
+
"epoch": 0.19,
|
2062 |
+
"grad_norm": 0.21995621919631958,
|
2063 |
+
"learning_rate": 9.994592234622619e-05,
|
2064 |
+
"loss": 2.5538,
|
2065 |
+
"step": 26400
|
2066 |
+
},
|
2067 |
+
{
|
2068 |
+
"epoch": 0.19,
|
2069 |
+
"grad_norm": 0.22063715755939484,
|
2070 |
+
"learning_rate": 9.99454781314916e-05,
|
2071 |
+
"loss": 2.5529,
|
2072 |
+
"step": 26500
|
2073 |
+
},
|
2074 |
+
{
|
2075 |
+
"epoch": 0.19,
|
2076 |
+
"grad_norm": 0.1817554086446762,
|
2077 |
+
"learning_rate": 9.99450321007318e-05,
|
2078 |
+
"loss": 2.5542,
|
2079 |
+
"step": 26600
|
2080 |
+
},
|
2081 |
+
{
|
2082 |
+
"epoch": 0.19,
|
2083 |
+
"grad_norm": 0.16820061206817627,
|
2084 |
+
"learning_rate": 9.994458425396298e-05,
|
2085 |
+
"loss": 2.5532,
|
2086 |
+
"step": 26700
|
2087 |
+
},
|
2088 |
+
{
|
2089 |
+
"epoch": 0.19,
|
2090 |
+
"grad_norm": 0.22539284825325012,
|
2091 |
+
"learning_rate": 9.994413459120146e-05,
|
2092 |
+
"loss": 2.5546,
|
2093 |
+
"step": 26800
|
2094 |
+
},
|
2095 |
+
{
|
2096 |
+
"epoch": 0.2,
|
2097 |
+
"grad_norm": 0.17431499063968658,
|
2098 |
+
"learning_rate": 9.994368311246356e-05,
|
2099 |
+
"loss": 2.5535,
|
2100 |
+
"step": 26900
|
2101 |
+
},
|
2102 |
+
{
|
2103 |
+
"epoch": 0.2,
|
2104 |
+
"grad_norm": 0.21042422950267792,
|
2105 |
+
"learning_rate": 9.99432298177657e-05,
|
2106 |
+
"loss": 2.554,
|
2107 |
+
"step": 27000
|
2108 |
+
},
|
2109 |
+
{
|
2110 |
+
"epoch": 0.2,
|
2111 |
+
"eval_loss": 2.5529098510742188,
|
2112 |
+
"eval_runtime": 7939.2552,
|
2113 |
+
"eval_samples_per_second": 738.741,
|
2114 |
+
"eval_steps_per_second": 46.171,
|
2115 |
+
"step": 27000
|
2116 |
+
},
|
2117 |
+
{
|
2118 |
+
"epoch": 0.2,
|
2119 |
+
"grad_norm": 0.18547575175762177,
|
2120 |
+
"learning_rate": 9.994277470712439e-05,
|
2121 |
+
"loss": 2.5533,
|
2122 |
+
"step": 27100
|
2123 |
+
},
|
2124 |
+
{
|
2125 |
+
"epoch": 0.2,
|
2126 |
+
"grad_norm": 0.17700788378715515,
|
2127 |
+
"learning_rate": 9.994231778055614e-05,
|
2128 |
+
"loss": 2.553,
|
2129 |
+
"step": 27200
|
2130 |
+
},
|
2131 |
+
{
|
2132 |
+
"epoch": 0.2,
|
2133 |
+
"grad_norm": 0.2162560373544693,
|
2134 |
+
"learning_rate": 9.99418590380776e-05,
|
2135 |
+
"loss": 2.5531,
|
2136 |
+
"step": 27300
|
2137 |
+
},
|
2138 |
+
{
|
2139 |
+
"epoch": 0.2,
|
2140 |
+
"grad_norm": 0.17806923389434814,
|
2141 |
+
"learning_rate": 9.99413984797054e-05,
|
2142 |
+
"loss": 2.5508,
|
2143 |
+
"step": 27400
|
2144 |
+
},
|
2145 |
+
{
|
2146 |
+
"epoch": 0.2,
|
2147 |
+
"grad_norm": 0.1959153264760971,
|
2148 |
+
"learning_rate": 9.994094073818738e-05,
|
2149 |
+
"loss": 2.5521,
|
2150 |
+
"step": 27500
|
2151 |
+
},
|
2152 |
+
{
|
2153 |
+
"epoch": 0.2,
|
2154 |
+
"grad_norm": 0.2098911553621292,
|
2155 |
+
"learning_rate": 9.994047656623675e-05,
|
2156 |
+
"loss": 2.5518,
|
2157 |
+
"step": 27600
|
2158 |
+
},
|
2159 |
+
{
|
2160 |
+
"epoch": 0.2,
|
2161 |
+
"grad_norm": 0.17420220375061035,
|
2162 |
+
"learning_rate": 9.994001057844278e-05,
|
2163 |
+
"loss": 2.5508,
|
2164 |
+
"step": 27700
|
2165 |
+
},
|
2166 |
+
{
|
2167 |
+
"epoch": 0.2,
|
2168 |
+
"grad_norm": 0.1538473218679428,
|
2169 |
+
"learning_rate": 9.993954277482238e-05,
|
2170 |
+
"loss": 2.5506,
|
2171 |
+
"step": 27800
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 0.2,
|
2175 |
+
"grad_norm": 0.1710011512041092,
|
2176 |
+
"learning_rate": 9.993907315539257e-05,
|
2177 |
+
"loss": 2.5509,
|
2178 |
+
"step": 27900
|
2179 |
+
},
|
2180 |
+
{
|
2181 |
+
"epoch": 0.2,
|
2182 |
+
"grad_norm": 0.14763277769088745,
|
2183 |
+
"learning_rate": 9.993860172017044e-05,
|
2184 |
+
"loss": 2.5514,
|
2185 |
+
"step": 28000
|
2186 |
+
},
|
2187 |
+
{
|
2188 |
+
"epoch": 0.2,
|
2189 |
+
"eval_loss": 2.5512099266052246,
|
2190 |
+
"eval_runtime": 9244.95,
|
2191 |
+
"eval_samples_per_second": 634.406,
|
2192 |
+
"eval_steps_per_second": 39.65,
|
2193 |
+
"step": 28000
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.2,
|
2197 |
+
"grad_norm": 0.24788980185985565,
|
2198 |
+
"learning_rate": 9.99381284691731e-05,
|
2199 |
+
"loss": 2.5507,
|
2200 |
+
"step": 28100
|
2201 |
+
},
|
2202 |
+
{
|
2203 |
+
"epoch": 0.21,
|
2204 |
+
"grad_norm": 0.16047827899456024,
|
2205 |
+
"learning_rate": 9.993765340241779e-05,
|
2206 |
+
"loss": 2.5504,
|
2207 |
+
"step": 28200
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 0.21,
|
2211 |
+
"grad_norm": 0.2097485363483429,
|
2212 |
+
"learning_rate": 9.993717651992176e-05,
|
2213 |
+
"loss": 2.5496,
|
2214 |
+
"step": 28300
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 0.21,
|
2218 |
+
"grad_norm": 0.15989455580711365,
|
2219 |
+
"learning_rate": 9.993669782170236e-05,
|
2220 |
+
"loss": 2.5511,
|
2221 |
+
"step": 28400
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 0.21,
|
2225 |
+
"grad_norm": 0.1990625262260437,
|
2226 |
+
"learning_rate": 9.9936217307777e-05,
|
2227 |
+
"loss": 2.5491,
|
2228 |
+
"step": 28500
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 0.21,
|
2232 |
+
"grad_norm": 0.17394055426120758,
|
2233 |
+
"learning_rate": 9.993573497816314e-05,
|
2234 |
+
"loss": 2.5503,
|
2235 |
+
"step": 28600
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 0.21,
|
2239 |
+
"grad_norm": 0.17688289284706116,
|
2240 |
+
"learning_rate": 9.993525083287832e-05,
|
2241 |
+
"loss": 2.5487,
|
2242 |
+
"step": 28700
|
2243 |
+
},
|
2244 |
+
{
|
2245 |
+
"epoch": 0.21,
|
2246 |
+
"grad_norm": 0.20647858083248138,
|
2247 |
+
"learning_rate": 9.993476487194015e-05,
|
2248 |
+
"loss": 2.5502,
|
2249 |
+
"step": 28800
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 0.21,
|
2253 |
+
"grad_norm": 0.17351941764354706,
|
2254 |
+
"learning_rate": 9.99342770953663e-05,
|
2255 |
+
"loss": 2.5503,
|
2256 |
+
"step": 28900
|
2257 |
+
},
|
2258 |
+
{
|
2259 |
+
"epoch": 0.21,
|
2260 |
+
"grad_norm": 0.16259269416332245,
|
2261 |
+
"learning_rate": 9.993379240808367e-05,
|
2262 |
+
"loss": 2.5505,
|
2263 |
+
"step": 29000
|
2264 |
+
},
|
2265 |
+
{
|
2266 |
+
"epoch": 0.21,
|
2267 |
+
"eval_loss": 2.5498733520507812,
|
2268 |
+
"eval_runtime": 8094.52,
|
2269 |
+
"eval_samples_per_second": 724.571,
|
2270 |
+
"eval_steps_per_second": 45.286,
|
2271 |
+
"step": 29000
|
2272 |
+
},
|
2273 |
+
{
|
2274 |
+
"epoch": 0.21,
|
2275 |
+
"grad_norm": 0.152243971824646,
|
2276 |
+
"learning_rate": 9.993330101844764e-05,
|
2277 |
+
"loss": 2.5494,
|
2278 |
+
"step": 29100
|
2279 |
+
},
|
2280 |
+
{
|
2281 |
+
"epoch": 0.21,
|
2282 |
+
"grad_norm": 0.21322308480739594,
|
2283 |
+
"learning_rate": 9.993280781322914e-05,
|
2284 |
+
"loss": 2.5495,
|
2285 |
+
"step": 29200
|
2286 |
+
},
|
2287 |
+
{
|
2288 |
+
"epoch": 0.21,
|
2289 |
+
"grad_norm": 0.17743776738643646,
|
2290 |
+
"learning_rate": 9.993231279244612e-05,
|
2291 |
+
"loss": 2.55,
|
2292 |
+
"step": 29300
|
2293 |
+
},
|
2294 |
+
{
|
2295 |
+
"epoch": 0.21,
|
2296 |
+
"grad_norm": 0.19075080752372742,
|
2297 |
+
"learning_rate": 9.993181595611659e-05,
|
2298 |
+
"loss": 2.5477,
|
2299 |
+
"step": 29400
|
2300 |
+
},
|
2301 |
+
{
|
2302 |
+
"epoch": 0.21,
|
2303 |
+
"grad_norm": 0.14956378936767578,
|
2304 |
+
"learning_rate": 9.993131730425858e-05,
|
2305 |
+
"loss": 2.5499,
|
2306 |
+
"step": 29500
|
2307 |
+
},
|
2308 |
+
{
|
2309 |
+
"epoch": 0.22,
|
2310 |
+
"grad_norm": 0.1913817673921585,
|
2311 |
+
"learning_rate": 9.993081683689026e-05,
|
2312 |
+
"loss": 2.5499,
|
2313 |
+
"step": 29600
|
2314 |
+
},
|
2315 |
+
{
|
2316 |
+
"epoch": 0.22,
|
2317 |
+
"grad_norm": 0.1898573786020279,
|
2318 |
+
"learning_rate": 9.99303145540298e-05,
|
2319 |
+
"loss": 2.5478,
|
2320 |
+
"step": 29700
|
2321 |
+
},
|
2322 |
+
{
|
2323 |
+
"epoch": 0.22,
|
2324 |
+
"grad_norm": 0.18223215639591217,
|
2325 |
+
"learning_rate": 9.992981045569545e-05,
|
2326 |
+
"loss": 2.5497,
|
2327 |
+
"step": 29800
|
2328 |
+
},
|
2329 |
+
{
|
2330 |
+
"epoch": 0.22,
|
2331 |
+
"grad_norm": 0.17807921767234802,
|
2332 |
+
"learning_rate": 9.992930454190558e-05,
|
2333 |
+
"loss": 2.5482,
|
2334 |
+
"step": 29900
|
2335 |
+
},
|
2336 |
+
{
|
2337 |
+
"epoch": 0.22,
|
2338 |
+
"grad_norm": 0.19245785474777222,
|
2339 |
+
"learning_rate": 9.992879681267855e-05,
|
2340 |
+
"loss": 2.5477,
|
2341 |
+
"step": 30000
|
2342 |
+
},
|
2343 |
+
{
|
2344 |
+
"epoch": 0.22,
|
2345 |
+
"eval_loss": 2.5480847358703613,
|
2346 |
+
"eval_runtime": 5418.7874,
|
2347 |
+
"eval_samples_per_second": 1082.355,
|
2348 |
+
"eval_steps_per_second": 67.647,
|
2349 |
+
"step": 30000
|
2350 |
+
},
|
2351 |
+
{
|
2352 |
+
"epoch": 0.22,
|
2353 |
+
"grad_norm": 0.16644108295440674,
|
2354 |
+
"learning_rate": 9.992828726803284e-05,
|
2355 |
+
"loss": 2.5462,
|
2356 |
+
"step": 30100
|
2357 |
+
},
|
2358 |
+
{
|
2359 |
+
"epoch": 0.22,
|
2360 |
+
"grad_norm": 0.21199798583984375,
|
2361 |
+
"learning_rate": 9.992777590798698e-05,
|
2362 |
+
"loss": 2.5472,
|
2363 |
+
"step": 30200
|
2364 |
+
},
|
2365 |
+
{
|
2366 |
+
"epoch": 0.22,
|
2367 |
+
"grad_norm": 0.16835100948810577,
|
2368 |
+
"learning_rate": 9.992726273255957e-05,
|
2369 |
+
"loss": 2.5492,
|
2370 |
+
"step": 30300
|
2371 |
+
},
|
2372 |
+
{
|
2373 |
+
"epoch": 0.22,
|
2374 |
+
"grad_norm": 0.24301642179489136,
|
2375 |
+
"learning_rate": 9.992674774176924e-05,
|
2376 |
+
"loss": 2.547,
|
2377 |
+
"step": 30400
|
2378 |
+
},
|
2379 |
+
{
|
2380 |
+
"epoch": 0.22,
|
2381 |
+
"grad_norm": 0.1855507344007492,
|
2382 |
+
"learning_rate": 9.992623093563473e-05,
|
2383 |
+
"loss": 2.5482,
|
2384 |
+
"step": 30500
|
2385 |
+
},
|
2386 |
+
{
|
2387 |
+
"epoch": 0.22,
|
2388 |
+
"grad_norm": 0.20825399458408356,
|
2389 |
+
"learning_rate": 9.992571231417482e-05,
|
2390 |
+
"loss": 2.5472,
|
2391 |
+
"step": 30600
|
2392 |
+
},
|
2393 |
+
{
|
2394 |
+
"epoch": 0.22,
|
2395 |
+
"grad_norm": 0.18983405828475952,
|
2396 |
+
"learning_rate": 9.99251918774084e-05,
|
2397 |
+
"loss": 2.5456,
|
2398 |
+
"step": 30700
|
2399 |
+
},
|
2400 |
+
{
|
2401 |
+
"epoch": 0.22,
|
2402 |
+
"grad_norm": 0.19868837296962738,
|
2403 |
+
"learning_rate": 9.992466962535437e-05,
|
2404 |
+
"loss": 2.5472,
|
2405 |
+
"step": 30800
|
2406 |
+
},
|
2407 |
+
{
|
2408 |
+
"epoch": 0.22,
|
2409 |
+
"grad_norm": 0.217344731092453,
|
2410 |
+
"learning_rate": 9.99241455580317e-05,
|
2411 |
+
"loss": 2.5468,
|
2412 |
+
"step": 30900
|
2413 |
+
},
|
2414 |
+
{
|
2415 |
+
"epoch": 0.23,
|
2416 |
+
"grad_norm": 0.21630564332008362,
|
2417 |
+
"learning_rate": 9.992361967545949e-05,
|
2418 |
+
"loss": 2.5461,
|
2419 |
+
"step": 31000
|
2420 |
+
},
|
2421 |
+
{
|
2422 |
+
"epoch": 0.23,
|
2423 |
+
"eval_loss": 2.5460619926452637,
|
2424 |
+
"eval_runtime": 5334.7258,
|
2425 |
+
"eval_samples_per_second": 1099.41,
|
2426 |
+
"eval_steps_per_second": 68.713,
|
2427 |
+
"step": 31000
|
2428 |
+
},
|
2429 |
+
{
|
2430 |
+
"epoch": 0.23,
|
2431 |
+
"grad_norm": 0.20726899802684784,
|
2432 |
+
"learning_rate": 9.992309197765681e-05,
|
2433 |
+
"loss": 2.5479,
|
2434 |
+
"step": 31100
|
2435 |
+
},
|
2436 |
+
{
|
2437 |
+
"epoch": 0.23,
|
2438 |
+
"grad_norm": 0.20710667967796326,
|
2439 |
+
"learning_rate": 9.99225624646429e-05,
|
2440 |
+
"loss": 2.5472,
|
2441 |
+
"step": 31200
|
2442 |
+
},
|
2443 |
+
{
|
2444 |
+
"epoch": 0.23,
|
2445 |
+
"grad_norm": 0.23984268307685852,
|
2446 |
+
"learning_rate": 9.992203113643699e-05,
|
2447 |
+
"loss": 2.5449,
|
2448 |
+
"step": 31300
|
2449 |
+
},
|
2450 |
+
{
|
2451 |
+
"epoch": 0.23,
|
2452 |
+
"grad_norm": 0.1859433650970459,
|
2453 |
+
"learning_rate": 9.992149799305838e-05,
|
2454 |
+
"loss": 2.5456,
|
2455 |
+
"step": 31400
|
2456 |
+
},
|
2457 |
+
{
|
2458 |
+
"epoch": 0.23,
|
2459 |
+
"grad_norm": 0.19020648300647736,
|
2460 |
+
"learning_rate": 9.992096303452647e-05,
|
2461 |
+
"loss": 2.5472,
|
2462 |
+
"step": 31500
|
2463 |
+
},
|
2464 |
+
{
|
2465 |
+
"epoch": 0.23,
|
2466 |
+
"grad_norm": 0.18619538843631744,
|
2467 |
+
"learning_rate": 9.992042626086073e-05,
|
2468 |
+
"loss": 2.5446,
|
2469 |
+
"step": 31600
|
2470 |
+
},
|
2471 |
+
{
|
2472 |
+
"epoch": 0.23,
|
2473 |
+
"grad_norm": 0.15984103083610535,
|
2474 |
+
"learning_rate": 9.991988767208065e-05,
|
2475 |
+
"loss": 2.5429,
|
2476 |
+
"step": 31700
|
2477 |
+
},
|
2478 |
+
{
|
2479 |
+
"epoch": 0.23,
|
2480 |
+
"grad_norm": 0.1874154657125473,
|
2481 |
+
"learning_rate": 9.991934726820583e-05,
|
2482 |
+
"loss": 2.5457,
|
2483 |
+
"step": 31800
|
2484 |
+
},
|
2485 |
+
{
|
2486 |
+
"epoch": 0.23,
|
2487 |
+
"grad_norm": 0.21573562920093536,
|
2488 |
+
"learning_rate": 9.99188050492559e-05,
|
2489 |
+
"loss": 2.5451,
|
2490 |
+
"step": 31900
|
2491 |
+
},
|
2492 |
+
{
|
2493 |
+
"epoch": 0.23,
|
2494 |
+
"grad_norm": 0.20766520500183105,
|
2495 |
+
"learning_rate": 9.991826101525059e-05,
|
2496 |
+
"loss": 2.5457,
|
2497 |
+
"step": 32000
|
2498 |
+
},
|
2499 |
+
{
|
2500 |
+
"epoch": 0.23,
|
2501 |
+
"eval_loss": 2.5444440841674805,
|
2502 |
+
"eval_runtime": 5469.9575,
|
2503 |
+
"eval_samples_per_second": 1072.23,
|
2504 |
+
"eval_steps_per_second": 67.014,
|
2505 |
+
"step": 32000
|
2506 |
+
},
|
2507 |
+
{
|
2508 |
+
"epoch": 0.23,
|
2509 |
+
"grad_norm": 0.1947011798620224,
|
2510 |
+
"learning_rate": 9.991771516620969e-05,
|
2511 |
+
"loss": 2.5439,
|
2512 |
+
"step": 32100
|
2513 |
+
},
|
2514 |
+
{
|
2515 |
+
"epoch": 0.23,
|
2516 |
+
"grad_norm": 0.20974138379096985,
|
2517 |
+
"learning_rate": 9.991716750215303e-05,
|
2518 |
+
"loss": 2.5455,
|
2519 |
+
"step": 32200
|
2520 |
+
},
|
2521 |
+
{
|
2522 |
+
"epoch": 0.23,
|
2523 |
+
"grad_norm": 0.20095574855804443,
|
2524 |
+
"learning_rate": 9.991661802310053e-05,
|
2525 |
+
"loss": 2.544,
|
2526 |
+
"step": 32300
|
2527 |
+
},
|
2528 |
+
{
|
2529 |
+
"epoch": 0.24,
|
2530 |
+
"grad_norm": 0.15232567489147186,
|
2531 |
+
"learning_rate": 9.991606672907218e-05,
|
2532 |
+
"loss": 2.5447,
|
2533 |
+
"step": 32400
|
2534 |
+
},
|
2535 |
+
{
|
2536 |
+
"epoch": 0.24,
|
2537 |
+
"grad_norm": 0.1890016496181488,
|
2538 |
+
"learning_rate": 9.99155191601618e-05,
|
2539 |
+
"loss": 2.5438,
|
2540 |
+
"step": 32500
|
2541 |
+
},
|
2542 |
+
{
|
2543 |
+
"epoch": 0.24,
|
2544 |
+
"grad_norm": 0.15099911391735077,
|
2545 |
+
"learning_rate": 9.991496425439117e-05,
|
2546 |
+
"loss": 2.5426,
|
2547 |
+
"step": 32600
|
2548 |
+
},
|
2549 |
+
{
|
2550 |
+
"epoch": 0.24,
|
2551 |
+
"grad_norm": 0.19564999639987946,
|
2552 |
+
"learning_rate": 9.991440753370483e-05,
|
2553 |
+
"loss": 2.5433,
|
2554 |
+
"step": 32700
|
2555 |
+
},
|
2556 |
+
{
|
2557 |
+
"epoch": 0.24,
|
2558 |
+
"grad_norm": 0.23213982582092285,
|
2559 |
+
"learning_rate": 9.991384899812299e-05,
|
2560 |
+
"loss": 2.5451,
|
2561 |
+
"step": 32800
|
2562 |
+
},
|
2563 |
+
{
|
2564 |
+
"epoch": 0.24,
|
2565 |
+
"grad_norm": 0.21088984608650208,
|
2566 |
+
"learning_rate": 9.991328864766595e-05,
|
2567 |
+
"loss": 2.5423,
|
2568 |
+
"step": 32900
|
2569 |
+
},
|
2570 |
+
{
|
2571 |
+
"epoch": 0.24,
|
2572 |
+
"grad_norm": 0.2080707848072052,
|
2573 |
+
"learning_rate": 9.991272648235412e-05,
|
2574 |
+
"loss": 2.542,
|
2575 |
+
"step": 33000
|
2576 |
+
},
|
2577 |
+
{
|
2578 |
+
"epoch": 0.24,
|
2579 |
+
"eval_loss": 2.543639898300171,
|
2580 |
+
"eval_runtime": 5500.5758,
|
2581 |
+
"eval_samples_per_second": 1066.262,
|
2582 |
+
"eval_steps_per_second": 66.641,
|
2583 |
+
"step": 33000
|
2584 |
+
},
|
2585 |
+
{
|
2586 |
+
"epoch": 0.24,
|
2587 |
+
"grad_norm": 0.20043937861919403,
|
2588 |
+
"learning_rate": 9.991216250220794e-05,
|
2589 |
+
"loss": 2.5434,
|
2590 |
+
"step": 33100
|
2591 |
+
},
|
2592 |
+
{
|
2593 |
+
"epoch": 0.24,
|
2594 |
+
"grad_norm": 0.26814088225364685,
|
2595 |
+
"learning_rate": 9.991159670724789e-05,
|
2596 |
+
"loss": 2.5439,
|
2597 |
+
"step": 33200
|
2598 |
+
},
|
2599 |
+
{
|
2600 |
+
"epoch": 0.24,
|
2601 |
+
"grad_norm": 0.1790783852338791,
|
2602 |
+
"learning_rate": 9.991102909749455e-05,
|
2603 |
+
"loss": 2.5422,
|
2604 |
+
"step": 33300
|
2605 |
+
},
|
2606 |
+
{
|
2607 |
+
"epoch": 0.24,
|
2608 |
+
"grad_norm": 0.17718929052352905,
|
2609 |
+
"learning_rate": 9.991045967296856e-05,
|
2610 |
+
"loss": 2.5414,
|
2611 |
+
"step": 33400
|
2612 |
+
},
|
2613 |
+
{
|
2614 |
+
"epoch": 0.24,
|
2615 |
+
"grad_norm": 0.19124653935432434,
|
2616 |
+
"learning_rate": 9.990988843369065e-05,
|
2617 |
+
"loss": 2.5409,
|
2618 |
+
"step": 33500
|
2619 |
+
},
|
2620 |
+
{
|
2621 |
+
"epoch": 0.24,
|
2622 |
+
"grad_norm": 0.19628183543682098,
|
2623 |
+
"learning_rate": 9.990931537968155e-05,
|
2624 |
+
"loss": 2.5423,
|
2625 |
+
"step": 33600
|
2626 |
+
},
|
2627 |
+
{
|
2628 |
+
"epoch": 0.25,
|
2629 |
+
"grad_norm": 0.2127145528793335,
|
2630 |
+
"learning_rate": 9.990874051096211e-05,
|
2631 |
+
"loss": 2.5451,
|
2632 |
+
"step": 33700
|
2633 |
+
},
|
2634 |
+
{
|
2635 |
+
"epoch": 0.25,
|
2636 |
+
"grad_norm": 0.19408264756202698,
|
2637 |
+
"learning_rate": 9.990816382755324e-05,
|
2638 |
+
"loss": 2.5405,
|
2639 |
+
"step": 33800
|
2640 |
+
},
|
2641 |
+
{
|
2642 |
+
"epoch": 0.25,
|
2643 |
+
"grad_norm": 0.1892174333333969,
|
2644 |
+
"learning_rate": 9.99075853294759e-05,
|
2645 |
+
"loss": 2.5428,
|
2646 |
+
"step": 33900
|
2647 |
+
},
|
2648 |
+
{
|
2649 |
+
"epoch": 0.25,
|
2650 |
+
"grad_norm": 0.20231932401657104,
|
2651 |
+
"learning_rate": 9.990700501675114e-05,
|
2652 |
+
"loss": 2.5424,
|
2653 |
+
"step": 34000
|
2654 |
+
},
|
2655 |
+
{
|
2656 |
+
"epoch": 0.25,
|
2657 |
+
"eval_loss": 2.5417420864105225,
|
2658 |
+
"eval_runtime": 5451.1377,
|
2659 |
+
"eval_samples_per_second": 1075.932,
|
2660 |
+
"eval_steps_per_second": 67.246,
|
2661 |
+
"step": 34000
|
2662 |
+
},
|
2663 |
+
{
|
2664 |
+
"epoch": 0.25,
|
2665 |
+
"grad_norm": 0.17818154394626617,
|
2666 |
+
"learning_rate": 9.990642288940005e-05,
|
2667 |
+
"loss": 2.5406,
|
2668 |
+
"step": 34100
|
2669 |
+
},
|
2670 |
+
{
|
2671 |
+
"epoch": 0.25,
|
2672 |
+
"grad_norm": 0.20383848249912262,
|
2673 |
+
"learning_rate": 9.990583894744378e-05,
|
2674 |
+
"loss": 2.5414,
|
2675 |
+
"step": 34200
|
2676 |
+
},
|
2677 |
+
{
|
2678 |
+
"epoch": 0.25,
|
2679 |
+
"grad_norm": 0.19464430212974548,
|
2680 |
+
"learning_rate": 9.99052531909036e-05,
|
2681 |
+
"loss": 2.5413,
|
2682 |
+
"step": 34300
|
2683 |
+
},
|
2684 |
+
{
|
2685 |
+
"epoch": 0.25,
|
2686 |
+
"grad_norm": 0.17793488502502441,
|
2687 |
+
"learning_rate": 9.990466561980076e-05,
|
2688 |
+
"loss": 2.5421,
|
2689 |
+
"step": 34400
|
2690 |
+
},
|
2691 |
+
{
|
2692 |
+
"epoch": 0.25,
|
2693 |
+
"grad_norm": 0.2119537591934204,
|
2694 |
+
"learning_rate": 9.990407623415668e-05,
|
2695 |
+
"loss": 2.54,
|
2696 |
+
"step": 34500
|
2697 |
+
},
|
2698 |
+
{
|
2699 |
+
"epoch": 0.25,
|
2700 |
+
"grad_norm": 0.20085354149341583,
|
2701 |
+
"learning_rate": 9.990348503399273e-05,
|
2702 |
+
"loss": 2.5395,
|
2703 |
+
"step": 34600
|
2704 |
+
},
|
2705 |
+
{
|
2706 |
+
"epoch": 0.25,
|
2707 |
+
"grad_norm": 0.18408524990081787,
|
2708 |
+
"learning_rate": 9.990289201933045e-05,
|
2709 |
+
"loss": 2.5395,
|
2710 |
+
"step": 34700
|
2711 |
+
},
|
2712 |
+
{
|
2713 |
+
"epoch": 0.25,
|
2714 |
+
"grad_norm": 0.18326416611671448,
|
2715 |
+
"learning_rate": 9.990229719019137e-05,
|
2716 |
+
"loss": 2.5404,
|
2717 |
+
"step": 34800
|
2718 |
+
},
|
2719 |
+
{
|
2720 |
+
"epoch": 0.25,
|
2721 |
+
"grad_norm": 0.2630787193775177,
|
2722 |
+
"learning_rate": 9.990170054659715e-05,
|
2723 |
+
"loss": 2.5404,
|
2724 |
+
"step": 34900
|
2725 |
+
},
|
2726 |
+
{
|
2727 |
+
"epoch": 0.25,
|
2728 |
+
"grad_norm": 0.23082153499126434,
|
2729 |
+
"learning_rate": 9.990110208856948e-05,
|
2730 |
+
"loss": 2.5402,
|
2731 |
+
"step": 35000
|
2732 |
+
},
|
2733 |
+
{
|
2734 |
+
"epoch": 0.25,
|
2735 |
+
"eval_loss": 2.540555477142334,
|
2736 |
+
"eval_runtime": 5286.1621,
|
2737 |
+
"eval_samples_per_second": 1109.511,
|
2738 |
+
"eval_steps_per_second": 69.344,
|
2739 |
+
"step": 35000
|
2740 |
+
},
|
2741 |
+
{
|
2742 |
+
"epoch": 0.26,
|
2743 |
+
"grad_norm": 0.2101873904466629,
|
2744 |
+
"learning_rate": 9.990050181613012e-05,
|
2745 |
+
"loss": 2.5409,
|
2746 |
+
"step": 35100
|
2747 |
+
},
|
2748 |
+
{
|
2749 |
+
"epoch": 0.26,
|
2750 |
+
"grad_norm": 0.1748313456773758,
|
2751 |
+
"learning_rate": 9.989989972930086e-05,
|
2752 |
+
"loss": 2.5408,
|
2753 |
+
"step": 35200
|
2754 |
+
},
|
2755 |
+
{
|
2756 |
+
"epoch": 0.26,
|
2757 |
+
"grad_norm": 0.21641358733177185,
|
2758 |
+
"learning_rate": 9.989929582810362e-05,
|
2759 |
+
"loss": 2.5407,
|
2760 |
+
"step": 35300
|
2761 |
+
},
|
2762 |
+
{
|
2763 |
+
"epoch": 0.26,
|
2764 |
+
"grad_norm": 0.20202140510082245,
|
2765 |
+
"learning_rate": 9.989869011256037e-05,
|
2766 |
+
"loss": 2.5419,
|
2767 |
+
"step": 35400
|
2768 |
+
},
|
2769 |
+
{
|
2770 |
+
"epoch": 0.26,
|
2771 |
+
"grad_norm": 0.1987978219985962,
|
2772 |
+
"learning_rate": 9.989808258269311e-05,
|
2773 |
+
"loss": 2.54,
|
2774 |
+
"step": 35500
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 0.26,
|
2778 |
+
"grad_norm": 0.17482729256153107,
|
2779 |
+
"learning_rate": 9.989747323852394e-05,
|
2780 |
+
"loss": 2.5398,
|
2781 |
+
"step": 35600
|
2782 |
+
},
|
2783 |
+
{
|
2784 |
+
"epoch": 0.26,
|
2785 |
+
"grad_norm": 0.15725675225257874,
|
2786 |
+
"learning_rate": 9.989686208007502e-05,
|
2787 |
+
"loss": 2.5384,
|
2788 |
+
"step": 35700
|
2789 |
+
},
|
2790 |
+
{
|
2791 |
+
"epoch": 0.26,
|
2792 |
+
"grad_norm": Infinity,
|
2793 |
+
"learning_rate": 9.989625524607613e-05,
|
2794 |
+
"loss": 2.54,
|
2795 |
+
"step": 35800
|
2796 |
+
},
|
2797 |
+
{
|
2798 |
+
"epoch": 0.26,
|
2799 |
+
"grad_norm": 0.19513057172298431,
|
2800 |
+
"learning_rate": 9.989564047727667e-05,
|
2801 |
+
"loss": 2.541,
|
2802 |
+
"step": 35900
|
2803 |
+
},
|
2804 |
+
{
|
2805 |
+
"epoch": 0.26,
|
2806 |
+
"grad_norm": 0.24705880880355835,
|
2807 |
+
"learning_rate": 9.989502389426411e-05,
|
2808 |
+
"loss": 2.5398,
|
2809 |
+
"step": 36000
|
2810 |
+
},
|
2811 |
+
{
|
2812 |
+
"epoch": 0.26,
|
2813 |
+
"eval_loss": 2.5393338203430176,
|
2814 |
+
"eval_runtime": 5370.7521,
|
2815 |
+
"eval_samples_per_second": 1092.036,
|
2816 |
+
"eval_steps_per_second": 68.252,
|
2817 |
+
"step": 36000
|
2818 |
+
},
|
2819 |
+
{
|
2820 |
+
"epoch": 0.26,
|
2821 |
+
"grad_norm": 0.17639793455600739,
|
2822 |
+
"learning_rate": 9.989440549706085e-05,
|
2823 |
+
"loss": 2.5398,
|
2824 |
+
"step": 36100
|
2825 |
+
},
|
2826 |
+
{
|
2827 |
+
"epoch": 0.26,
|
2828 |
+
"grad_norm": 0.23090311884880066,
|
2829 |
+
"learning_rate": 9.989378528568935e-05,
|
2830 |
+
"loss": 2.5399,
|
2831 |
+
"step": 36200
|
2832 |
+
},
|
2833 |
+
{
|
2834 |
+
"epoch": 0.26,
|
2835 |
+
"grad_norm": 0.21377325057983398,
|
2836 |
+
"learning_rate": 9.989316326017221e-05,
|
2837 |
+
"loss": 2.5351,
|
2838 |
+
"step": 36300
|
2839 |
+
},
|
2840 |
+
{
|
2841 |
+
"epoch": 0.26,
|
2842 |
+
"grad_norm": 0.21422795951366425,
|
2843 |
+
"learning_rate": 9.989253942053204e-05,
|
2844 |
+
"loss": 2.5381,
|
2845 |
+
"step": 36400
|
2846 |
+
},
|
2847 |
+
{
|
2848 |
+
"epoch": 0.27,
|
2849 |
+
"grad_norm": 0.20974934101104736,
|
2850 |
+
"learning_rate": 9.98919137667915e-05,
|
2851 |
+
"loss": 2.5378,
|
2852 |
+
"step": 36500
|
2853 |
+
},
|
2854 |
+
{
|
2855 |
+
"epoch": 0.27,
|
2856 |
+
"grad_norm": 0.19352389872074127,
|
2857 |
+
"learning_rate": 9.989128629897335e-05,
|
2858 |
+
"loss": 2.5378,
|
2859 |
+
"step": 36600
|
2860 |
+
},
|
2861 |
+
{
|
2862 |
+
"epoch": 0.27,
|
2863 |
+
"grad_norm": 0.21016819775104523,
|
2864 |
+
"learning_rate": 9.989065701710041e-05,
|
2865 |
+
"loss": 2.5366,
|
2866 |
+
"step": 36700
|
2867 |
+
},
|
2868 |
+
{
|
2869 |
+
"epoch": 0.27,
|
2870 |
+
"grad_norm": 0.1750701367855072,
|
2871 |
+
"learning_rate": 9.989002592119554e-05,
|
2872 |
+
"loss": 2.5399,
|
2873 |
+
"step": 36800
|
2874 |
+
},
|
2875 |
+
{
|
2876 |
+
"epoch": 0.27,
|
2877 |
+
"grad_norm": 0.18955004215240479,
|
2878 |
+
"learning_rate": 9.988939301128171e-05,
|
2879 |
+
"loss": 2.5411,
|
2880 |
+
"step": 36900
|
2881 |
+
},
|
2882 |
+
{
|
2883 |
+
"epoch": 0.27,
|
2884 |
+
"grad_norm": 0.22183337807655334,
|
2885 |
+
"learning_rate": 9.988875828738192e-05,
|
2886 |
+
"loss": 2.5385,
|
2887 |
+
"step": 37000
|
2888 |
+
},
|
2889 |
+
{
|
2890 |
+
"epoch": 0.27,
|
2891 |
+
"eval_loss": 2.537937879562378,
|
2892 |
+
"eval_runtime": 5267.1428,
|
2893 |
+
"eval_samples_per_second": 1113.517,
|
2894 |
+
"eval_steps_per_second": 69.595,
|
2895 |
+
"step": 37000
|
2896 |
+
},
|
2897 |
+
{
|
2898 |
+
"epoch": 0.27,
|
2899 |
+
"grad_norm": 0.17703290283679962,
|
2900 |
+
"learning_rate": 9.988812174951926e-05,
|
2901 |
+
"loss": 2.5386,
|
2902 |
+
"step": 37100
|
2903 |
+
},
|
2904 |
+
{
|
2905 |
+
"epoch": 0.27,
|
2906 |
+
"grad_norm": 0.17456910014152527,
|
2907 |
+
"learning_rate": 9.988748339771686e-05,
|
2908 |
+
"loss": 2.536,
|
2909 |
+
"step": 37200
|
2910 |
+
},
|
2911 |
+
{
|
2912 |
+
"epoch": 0.27,
|
2913 |
+
"grad_norm": 0.1729470044374466,
|
2914 |
+
"learning_rate": 9.988684323199795e-05,
|
2915 |
+
"loss": 2.5367,
|
2916 |
+
"step": 37300
|
2917 |
+
},
|
2918 |
+
{
|
2919 |
+
"epoch": 0.27,
|
2920 |
+
"grad_norm": 0.14801037311553955,
|
2921 |
+
"learning_rate": 9.988620125238578e-05,
|
2922 |
+
"loss": 2.5365,
|
2923 |
+
"step": 37400
|
2924 |
+
},
|
2925 |
+
{
|
2926 |
+
"epoch": 0.27,
|
2927 |
+
"grad_norm": 0.19737888872623444,
|
2928 |
+
"learning_rate": 9.988555745890371e-05,
|
2929 |
+
"loss": 2.5373,
|
2930 |
+
"step": 37500
|
2931 |
+
},
|
2932 |
+
{
|
2933 |
+
"epoch": 0.27,
|
2934 |
+
"grad_norm": 0.2556416392326355,
|
2935 |
+
"learning_rate": 9.988491185157514e-05,
|
2936 |
+
"loss": 2.5389,
|
2937 |
+
"step": 37600
|
2938 |
+
},
|
2939 |
+
{
|
2940 |
+
"epoch": 0.27,
|
2941 |
+
"grad_norm": 0.22999688982963562,
|
2942 |
+
"learning_rate": 9.988426443042357e-05,
|
2943 |
+
"loss": 2.5366,
|
2944 |
+
"step": 37700
|
2945 |
+
},
|
2946 |
+
{
|
2947 |
+
"epoch": 0.27,
|
2948 |
+
"grad_norm": 0.20399746298789978,
|
2949 |
+
"learning_rate": 9.988361519547252e-05,
|
2950 |
+
"loss": 2.5375,
|
2951 |
+
"step": 37800
|
2952 |
+
},
|
2953 |
+
{
|
2954 |
+
"epoch": 0.28,
|
2955 |
+
"grad_norm": 0.1886683702468872,
|
2956 |
+
"learning_rate": 9.988296414674556e-05,
|
2957 |
+
"loss": 2.5378,
|
2958 |
+
"step": 37900
|
2959 |
+
},
|
2960 |
+
{
|
2961 |
+
"epoch": 0.28,
|
2962 |
+
"grad_norm": 0.15870767831802368,
|
2963 |
+
"learning_rate": 9.988231128426643e-05,
|
2964 |
+
"loss": 2.5396,
|
2965 |
+
"step": 38000
|
2966 |
+
},
|
2967 |
+
{
|
2968 |
+
"epoch": 0.28,
|
2969 |
+
"eval_loss": 2.536235809326172,
|
2970 |
+
"eval_runtime": 5469.5004,
|
2971 |
+
"eval_samples_per_second": 1072.32,
|
2972 |
+
"eval_steps_per_second": 67.02,
|
2973 |
+
"step": 38000
|
2974 |
+
},
|
2975 |
+
{
|
2976 |
+
"epoch": 0.28,
|
2977 |
+
"grad_norm": 0.19297446310520172,
|
2978 |
+
"learning_rate": 9.988165660805883e-05,
|
2979 |
+
"loss": 2.5364,
|
2980 |
+
"step": 38100
|
2981 |
+
},
|
2982 |
+
{
|
2983 |
+
"epoch": 0.28,
|
2984 |
+
"grad_norm": 0.1729346513748169,
|
2985 |
+
"learning_rate": 9.988100011814657e-05,
|
2986 |
+
"loss": 2.5374,
|
2987 |
+
"step": 38200
|
2988 |
+
},
|
2989 |
+
{
|
2990 |
+
"epoch": 0.28,
|
2991 |
+
"grad_norm": 0.2072724550962448,
|
2992 |
+
"learning_rate": 9.988034181455352e-05,
|
2993 |
+
"loss": 2.5381,
|
2994 |
+
"step": 38300
|
2995 |
+
},
|
2996 |
+
{
|
2997 |
+
"epoch": 0.28,
|
2998 |
+
"grad_norm": 0.1756744384765625,
|
2999 |
+
"learning_rate": 9.987968169730362e-05,
|
3000 |
+
"loss": 2.5379,
|
3001 |
+
"step": 38400
|
3002 |
+
},
|
3003 |
+
{
|
3004 |
+
"epoch": 0.28,
|
3005 |
+
"grad_norm": 0.27329882979393005,
|
3006 |
+
"learning_rate": 9.987901976642086e-05,
|
3007 |
+
"loss": 2.5349,
|
3008 |
+
"step": 38500
|
3009 |
+
},
|
3010 |
+
{
|
3011 |
+
"epoch": 0.28,
|
3012 |
+
"grad_norm": 0.23222842812538147,
|
3013 |
+
"learning_rate": 9.987835602192934e-05,
|
3014 |
+
"loss": 2.5355,
|
3015 |
+
"step": 38600
|
3016 |
+
},
|
3017 |
+
{
|
3018 |
+
"epoch": 0.28,
|
3019 |
+
"grad_norm": 0.2461015284061432,
|
3020 |
+
"learning_rate": 9.987769046385316e-05,
|
3021 |
+
"loss": 2.5364,
|
3022 |
+
"step": 38700
|
3023 |
+
},
|
3024 |
+
{
|
3025 |
+
"epoch": 0.28,
|
3026 |
+
"grad_norm": 0.1758180409669876,
|
3027 |
+
"learning_rate": 9.987702309221651e-05,
|
3028 |
+
"loss": 2.5377,
|
3029 |
+
"step": 38800
|
3030 |
+
},
|
3031 |
+
{
|
3032 |
+
"epoch": 0.28,
|
3033 |
+
"grad_norm": 0.15205176174640656,
|
3034 |
+
"learning_rate": 9.987635390704369e-05,
|
3035 |
+
"loss": 2.5376,
|
3036 |
+
"step": 38900
|
3037 |
+
},
|
3038 |
+
{
|
3039 |
+
"epoch": 0.28,
|
3040 |
+
"grad_norm": 0.1985970139503479,
|
3041 |
+
"learning_rate": 9.987568290835903e-05,
|
3042 |
+
"loss": 2.5372,
|
3043 |
+
"step": 39000
|
3044 |
+
},
|
3045 |
+
{
|
3046 |
+
"epoch": 0.28,
|
3047 |
+
"eval_loss": 2.53525710105896,
|
3048 |
+
"eval_runtime": 5498.4571,
|
3049 |
+
"eval_samples_per_second": 1066.673,
|
3050 |
+
"eval_steps_per_second": 66.667,
|
3051 |
+
"step": 39000
|
3052 |
+
},
|
3053 |
+
{
|
3054 |
+
"epoch": 0.28,
|
3055 |
+
"grad_norm": 0.1800653338432312,
|
3056 |
+
"learning_rate": 9.987501009618691e-05,
|
3057 |
+
"loss": 2.5349,
|
3058 |
+
"step": 39100
|
3059 |
+
},
|
3060 |
+
{
|
3061 |
+
"epoch": 0.29,
|
3062 |
+
"grad_norm": 0.22484809160232544,
|
3063 |
+
"learning_rate": 9.987433547055178e-05,
|
3064 |
+
"loss": 2.5364,
|
3065 |
+
"step": 39200
|
3066 |
+
},
|
3067 |
+
{
|
3068 |
+
"epoch": 0.29,
|
3069 |
+
"grad_norm": 0.1974021941423416,
|
3070 |
+
"learning_rate": 9.98736590314782e-05,
|
3071 |
+
"loss": 2.5333,
|
3072 |
+
"step": 39300
|
3073 |
+
},
|
3074 |
+
{
|
3075 |
+
"epoch": 0.29,
|
3076 |
+
"grad_norm": 0.23238864541053772,
|
3077 |
+
"learning_rate": 9.987298077899076e-05,
|
3078 |
+
"loss": 2.5371,
|
3079 |
+
"step": 39400
|
3080 |
+
},
|
3081 |
+
{
|
3082 |
+
"epoch": 0.29,
|
3083 |
+
"grad_norm": 0.17493529617786407,
|
3084 |
+
"learning_rate": 9.987230071311411e-05,
|
3085 |
+
"loss": 2.5356,
|
3086 |
+
"step": 39500
|
3087 |
+
},
|
3088 |
+
{
|
3089 |
+
"epoch": 0.29,
|
3090 |
+
"grad_norm": 0.23257118463516235,
|
3091 |
+
"learning_rate": 9.987161883387299e-05,
|
3092 |
+
"loss": 2.5354,
|
3093 |
+
"step": 39600
|
3094 |
+
},
|
3095 |
+
{
|
3096 |
+
"epoch": 0.29,
|
3097 |
+
"grad_norm": 0.18204239010810852,
|
3098 |
+
"learning_rate": 9.987094198719394e-05,
|
3099 |
+
"loss": 2.5348,
|
3100 |
+
"step": 39700
|
3101 |
+
},
|
3102 |
+
{
|
3103 |
+
"epoch": 0.29,
|
3104 |
+
"grad_norm": 0.20585016906261444,
|
3105 |
+
"learning_rate": 9.987025649943133e-05,
|
3106 |
+
"loss": 2.5347,
|
3107 |
+
"step": 39800
|
3108 |
+
},
|
3109 |
+
{
|
3110 |
+
"epoch": 0.29,
|
3111 |
+
"grad_norm": 0.19951903820037842,
|
3112 |
+
"learning_rate": 9.986956919837858e-05,
|
3113 |
+
"loss": 2.5355,
|
3114 |
+
"step": 39900
|
3115 |
+
},
|
3116 |
+
{
|
3117 |
+
"epoch": 0.29,
|
3118 |
+
"grad_norm": 0.1689595878124237,
|
3119 |
+
"learning_rate": 9.986888008406065e-05,
|
3120 |
+
"loss": 2.535,
|
3121 |
+
"step": 40000
|
3122 |
+
},
|
3123 |
+
{
|
3124 |
+
"epoch": 0.29,
|
3125 |
+
"eval_loss": 2.534259796142578,
|
3126 |
+
"eval_runtime": 5327.044,
|
3127 |
+
"eval_samples_per_second": 1100.996,
|
3128 |
+
"eval_steps_per_second": 68.812,
|
3129 |
+
"step": 40000
|
3130 |
+
},
|
3131 |
+
{
|
3132 |
+
"epoch": 0.29,
|
3133 |
+
"grad_norm": 0.16131243109703064,
|
3134 |
+
"learning_rate": 9.986818915650265e-05,
|
3135 |
+
"loss": 2.5343,
|
3136 |
+
"step": 40100
|
3137 |
+
},
|
3138 |
+
{
|
3139 |
+
"epoch": 0.29,
|
3140 |
+
"grad_norm": 0.16442734003067017,
|
3141 |
+
"learning_rate": 9.986749641572963e-05,
|
3142 |
+
"loss": 2.5336,
|
3143 |
+
"step": 40200
|
3144 |
+
},
|
3145 |
+
{
|
3146 |
+
"epoch": 0.29,
|
3147 |
+
"grad_norm": 0.17911262810230255,
|
3148 |
+
"learning_rate": 9.986680186176684e-05,
|
3149 |
+
"loss": 2.534,
|
3150 |
+
"step": 40300
|
3151 |
+
},
|
3152 |
+
{
|
3153 |
+
"epoch": 0.29,
|
3154 |
+
"grad_norm": 0.19391265511512756,
|
3155 |
+
"learning_rate": 9.986610549463952e-05,
|
3156 |
+
"loss": 2.5344,
|
3157 |
+
"step": 40400
|
3158 |
+
},
|
3159 |
+
{
|
3160 |
+
"epoch": 0.29,
|
3161 |
+
"grad_norm": 0.21224987506866455,
|
3162 |
+
"learning_rate": 9.986540731437298e-05,
|
3163 |
+
"loss": 2.5362,
|
3164 |
+
"step": 40500
|
3165 |
+
},
|
3166 |
+
{
|
3167 |
+
"epoch": 0.3,
|
3168 |
+
"grad_norm": 0.21114754676818848,
|
3169 |
+
"learning_rate": 9.986470732099258e-05,
|
3170 |
+
"loss": 2.5344,
|
3171 |
+
"step": 40600
|
3172 |
+
},
|
3173 |
+
{
|
3174 |
+
"epoch": 0.3,
|
3175 |
+
"grad_norm": 0.17994599044322968,
|
3176 |
+
"learning_rate": 9.986400551452382e-05,
|
3177 |
+
"loss": 2.5338,
|
3178 |
+
"step": 40700
|
3179 |
+
},
|
3180 |
+
{
|
3181 |
+
"epoch": 0.3,
|
3182 |
+
"grad_norm": 0.20839715003967285,
|
3183 |
+
"learning_rate": 9.98633018949922e-05,
|
3184 |
+
"loss": 2.5327,
|
3185 |
+
"step": 40800
|
3186 |
+
},
|
3187 |
+
{
|
3188 |
+
"epoch": 0.3,
|
3189 |
+
"grad_norm": 0.1808551549911499,
|
3190 |
+
"learning_rate": 9.986259646242329e-05,
|
3191 |
+
"loss": 2.5323,
|
3192 |
+
"step": 40900
|
3193 |
+
},
|
3194 |
+
{
|
3195 |
+
"epoch": 0.3,
|
3196 |
+
"grad_norm": 0.20471493899822235,
|
3197 |
+
"learning_rate": 9.986188921684276e-05,
|
3198 |
+
"loss": 2.5318,
|
3199 |
+
"step": 41000
|
3200 |
+
},
|
3201 |
+
{
|
3202 |
+
"epoch": 0.3,
|
3203 |
+
"eval_loss": 2.5325772762298584,
|
3204 |
+
"eval_runtime": 5462.6174,
|
3205 |
+
"eval_samples_per_second": 1073.671,
|
3206 |
+
"eval_steps_per_second": 67.104,
|
3207 |
+
"step": 41000
|
3208 |
+
},
|
3209 |
+
{
|
3210 |
+
"epoch": 0.3,
|
3211 |
+
"grad_norm": 0.18955881893634796,
|
3212 |
+
"learning_rate": 9.986118015827632e-05,
|
3213 |
+
"loss": 2.5328,
|
3214 |
+
"step": 41100
|
3215 |
+
},
|
3216 |
+
{
|
3217 |
+
"epoch": 0.3,
|
3218 |
+
"grad_norm": 0.20864000916481018,
|
3219 |
+
"learning_rate": 9.986046928674974e-05,
|
3220 |
+
"loss": 2.5303,
|
3221 |
+
"step": 41200
|
3222 |
+
},
|
3223 |
+
{
|
3224 |
+
"epoch": 0.3,
|
3225 |
+
"grad_norm": 0.17366698384284973,
|
3226 |
+
"learning_rate": 9.985975660228888e-05,
|
3227 |
+
"loss": 2.5314,
|
3228 |
+
"step": 41300
|
3229 |
+
},
|
3230 |
+
{
|
3231 |
+
"epoch": 0.3,
|
3232 |
+
"grad_norm": 0.2348269373178482,
|
3233 |
+
"learning_rate": 9.985904210491963e-05,
|
3234 |
+
"loss": 2.5336,
|
3235 |
+
"step": 41400
|
3236 |
+
},
|
3237 |
+
{
|
3238 |
+
"epoch": 0.3,
|
3239 |
+
"grad_norm": 0.2203817069530487,
|
3240 |
+
"learning_rate": 9.985832579466801e-05,
|
3241 |
+
"loss": 2.5307,
|
3242 |
+
"step": 41500
|
3243 |
+
},
|
3244 |
+
{
|
3245 |
+
"epoch": 0.3,
|
3246 |
+
"grad_norm": 0.20355217158794403,
|
3247 |
+
"learning_rate": 9.985760767156003e-05,
|
3248 |
+
"loss": 2.5329,
|
3249 |
+
"step": 41600
|
3250 |
+
},
|
3251 |
+
{
|
3252 |
+
"epoch": 0.3,
|
3253 |
+
"grad_norm": 0.21894210577011108,
|
3254 |
+
"learning_rate": 9.985688773562183e-05,
|
3255 |
+
"loss": 2.5325,
|
3256 |
+
"step": 41700
|
3257 |
+
},
|
3258 |
+
{
|
3259 |
+
"epoch": 0.3,
|
3260 |
+
"grad_norm": 0.18119728565216064,
|
3261 |
+
"learning_rate": 9.985616598687954e-05,
|
3262 |
+
"loss": 2.5316,
|
3263 |
+
"step": 41800
|
3264 |
+
},
|
3265 |
+
{
|
3266 |
+
"epoch": 0.3,
|
3267 |
+
"grad_norm": 0.22831158339977264,
|
3268 |
+
"learning_rate": 9.985544242535946e-05,
|
3269 |
+
"loss": 2.5316,
|
3270 |
+
"step": 41900
|
3271 |
+
},
|
3272 |
+
{
|
3273 |
+
"epoch": 0.31,
|
3274 |
+
"grad_norm": 0.18714164197444916,
|
3275 |
+
"learning_rate": 9.985471705108787e-05,
|
3276 |
+
"loss": 2.5315,
|
3277 |
+
"step": 42000
|
3278 |
+
},
|
3279 |
+
{
|
3280 |
+
"epoch": 0.31,
|
3281 |
+
"eval_loss": 2.531712770462036,
|
3282 |
+
"eval_runtime": 5461.1677,
|
3283 |
+
"eval_samples_per_second": 1073.956,
|
3284 |
+
"eval_steps_per_second": 67.122,
|
3285 |
+
"step": 42000
|
3286 |
+
},
|
3287 |
+
{
|
3288 |
+
"epoch": 0.31,
|
3289 |
+
"grad_norm": 0.19056876003742218,
|
3290 |
+
"learning_rate": 9.985398986409115e-05,
|
3291 |
+
"loss": 2.5305,
|
3292 |
+
"step": 42100
|
3293 |
+
},
|
3294 |
+
{
|
3295 |
+
"epoch": 0.31,
|
3296 |
+
"grad_norm": 0.19371190667152405,
|
3297 |
+
"learning_rate": 9.985326086439573e-05,
|
3298 |
+
"loss": 2.5333,
|
3299 |
+
"step": 42200
|
3300 |
+
},
|
3301 |
+
{
|
3302 |
+
"epoch": 0.31,
|
3303 |
+
"grad_norm": 0.19493506848812103,
|
3304 |
+
"learning_rate": 9.985253005202813e-05,
|
3305 |
+
"loss": 2.5279,
|
3306 |
+
"step": 42300
|
3307 |
+
},
|
3308 |
+
{
|
3309 |
+
"epoch": 0.31,
|
3310 |
+
"grad_norm": 0.17453902959823608,
|
3311 |
+
"learning_rate": 9.985179742701491e-05,
|
3312 |
+
"loss": 2.5334,
|
3313 |
+
"step": 42400
|
3314 |
+
},
|
3315 |
+
{
|
3316 |
+
"epoch": 0.31,
|
3317 |
+
"grad_norm": 0.19747483730316162,
|
3318 |
+
"learning_rate": 9.985106298938272e-05,
|
3319 |
+
"loss": 2.5321,
|
3320 |
+
"step": 42500
|
3321 |
+
},
|
3322 |
+
{
|
3323 |
+
"epoch": 0.31,
|
3324 |
+
"grad_norm": 0.18142738938331604,
|
3325 |
+
"learning_rate": 9.985032673915826e-05,
|
3326 |
+
"loss": 2.5312,
|
3327 |
+
"step": 42600
|
3328 |
+
},
|
3329 |
+
{
|
3330 |
+
"epoch": 0.31,
|
3331 |
+
"grad_norm": 0.21015243232250214,
|
3332 |
+
"learning_rate": 9.984958867636828e-05,
|
3333 |
+
"loss": 2.5328,
|
3334 |
+
"step": 42700
|
3335 |
+
},
|
3336 |
+
{
|
3337 |
+
"epoch": 0.31,
|
3338 |
+
"grad_norm": 0.20541459321975708,
|
3339 |
+
"learning_rate": 9.984884880103966e-05,
|
3340 |
+
"loss": 2.5318,
|
3341 |
+
"step": 42800
|
3342 |
+
},
|
3343 |
+
{
|
3344 |
+
"epoch": 0.31,
|
3345 |
+
"grad_norm": 0.24034422636032104,
|
3346 |
+
"learning_rate": 9.984810711319927e-05,
|
3347 |
+
"loss": 2.5305,
|
3348 |
+
"step": 42900
|
3349 |
+
},
|
3350 |
+
{
|
3351 |
+
"epoch": 0.31,
|
3352 |
+
"grad_norm": 0.22423624992370605,
|
3353 |
+
"learning_rate": 9.98473636128741e-05,
|
3354 |
+
"loss": 2.5298,
|
3355 |
+
"step": 43000
|
3356 |
+
},
|
3357 |
+
{
|
3358 |
+
"epoch": 0.31,
|
3359 |
+
"eval_loss": 2.530484437942505,
|
3360 |
+
"eval_runtime": 5481.4821,
|
3361 |
+
"eval_samples_per_second": 1069.976,
|
3362 |
+
"eval_steps_per_second": 66.874,
|
3363 |
+
"step": 43000
|
3364 |
+
},
|
3365 |
+
{
|
3366 |
+
"epoch": 0.31,
|
3367 |
+
"grad_norm": 0.1877780258655548,
|
3368 |
+
"learning_rate": 9.984661830009115e-05,
|
3369 |
+
"loss": 2.5308,
|
3370 |
+
"step": 43100
|
3371 |
+
},
|
3372 |
+
{
|
3373 |
+
"epoch": 0.31,
|
3374 |
+
"grad_norm": 0.2150457501411438,
|
3375 |
+
"learning_rate": 9.984587117487755e-05,
|
3376 |
+
"loss": 2.5318,
|
3377 |
+
"step": 43200
|
3378 |
+
},
|
3379 |
+
{
|
3380 |
+
"epoch": 0.31,
|
3381 |
+
"grad_norm": 0.17595185339450836,
|
3382 |
+
"learning_rate": 9.984512223726045e-05,
|
3383 |
+
"loss": 2.5291,
|
3384 |
+
"step": 43300
|
3385 |
+
},
|
3386 |
+
{
|
3387 |
+
"epoch": 0.32,
|
3388 |
+
"grad_norm": 0.1872701346874237,
|
3389 |
+
"learning_rate": 9.984437148726708e-05,
|
3390 |
+
"loss": 2.5298,
|
3391 |
+
"step": 43400
|
3392 |
+
},
|
3393 |
+
{
|
3394 |
+
"epoch": 0.32,
|
3395 |
+
"grad_norm": 0.15969082713127136,
|
3396 |
+
"learning_rate": 9.984361892492475e-05,
|
3397 |
+
"loss": 2.5322,
|
3398 |
+
"step": 43500
|
3399 |
+
},
|
3400 |
+
{
|
3401 |
+
"epoch": 0.32,
|
3402 |
+
"grad_norm": 0.22320838272571564,
|
3403 |
+
"learning_rate": 9.984286455026082e-05,
|
3404 |
+
"loss": 2.5284,
|
3405 |
+
"step": 43600
|
3406 |
+
},
|
3407 |
+
{
|
3408 |
+
"epoch": 0.32,
|
3409 |
+
"grad_norm": 0.22105048596858978,
|
3410 |
+
"learning_rate": 9.984211593414306e-05,
|
3411 |
+
"loss": 2.529,
|
3412 |
+
"step": 43700
|
3413 |
+
},
|
3414 |
+
{
|
3415 |
+
"epoch": 0.32,
|
3416 |
+
"grad_norm": 0.18613219261169434,
|
3417 |
+
"learning_rate": 9.98413579530408e-05,
|
3418 |
+
"loss": 2.5306,
|
3419 |
+
"step": 43800
|
3420 |
+
},
|
3421 |
+
{
|
3422 |
+
"epoch": 0.32,
|
3423 |
+
"grad_norm": 0.21678748726844788,
|
3424 |
+
"learning_rate": 9.984059815969915e-05,
|
3425 |
+
"loss": 2.5296,
|
3426 |
+
"step": 43900
|
3427 |
+
},
|
3428 |
+
{
|
3429 |
+
"epoch": 0.32,
|
3430 |
+
"grad_norm": 0.19831928610801697,
|
3431 |
+
"learning_rate": 9.983983655414574e-05,
|
3432 |
+
"loss": 2.5296,
|
3433 |
+
"step": 44000
|
3434 |
+
},
|
3435 |
+
{
|
3436 |
+
"epoch": 0.32,
|
3437 |
+
"eval_loss": 2.5292155742645264,
|
3438 |
+
"eval_runtime": 5508.261,
|
3439 |
+
"eval_samples_per_second": 1064.774,
|
3440 |
+
"eval_steps_per_second": 66.548,
|
3441 |
+
"step": 44000
|
3442 |
+
},
|
3443 |
+
{
|
3444 |
+
"epoch": 0.32,
|
3445 |
+
"grad_norm": NaN,
|
3446 |
+
"learning_rate": 9.983908077955583e-05,
|
3447 |
+
"loss": 2.5294,
|
3448 |
+
"step": 44100
|
3449 |
+
},
|
3450 |
+
{
|
3451 |
+
"epoch": 0.32,
|
3452 |
+
"grad_norm": 0.18271589279174805,
|
3453 |
+
"learning_rate": 9.983831556778345e-05,
|
3454 |
+
"loss": 2.5312,
|
3455 |
+
"step": 44200
|
3456 |
+
},
|
3457 |
+
{
|
3458 |
+
"epoch": 0.32,
|
3459 |
+
"grad_norm": 0.24201270937919617,
|
3460 |
+
"learning_rate": 9.98375485438823e-05,
|
3461 |
+
"loss": 2.5303,
|
3462 |
+
"step": 44300
|
3463 |
+
},
|
3464 |
+
{
|
3465 |
+
"epoch": 0.32,
|
3466 |
+
"grad_norm": 0.18819460272789001,
|
3467 |
+
"learning_rate": 9.983677970788026e-05,
|
3468 |
+
"loss": 2.5285,
|
3469 |
+
"step": 44400
|
3470 |
+
},
|
3471 |
+
{
|
3472 |
+
"epoch": 0.32,
|
3473 |
+
"grad_norm": 0.16757667064666748,
|
3474 |
+
"learning_rate": 9.983600905980529e-05,
|
3475 |
+
"loss": 2.5289,
|
3476 |
+
"step": 44500
|
3477 |
+
},
|
3478 |
+
{
|
3479 |
+
"epoch": 0.32,
|
3480 |
+
"grad_norm": 0.16967040300369263,
|
3481 |
+
"learning_rate": 9.983523659968541e-05,
|
3482 |
+
"loss": 2.5306,
|
3483 |
+
"step": 44600
|
3484 |
+
},
|
3485 |
+
{
|
3486 |
+
"epoch": 0.33,
|
3487 |
+
"grad_norm": 0.19566090404987335,
|
3488 |
+
"learning_rate": 9.983446232754872e-05,
|
3489 |
+
"loss": 2.5276,
|
3490 |
+
"step": 44700
|
3491 |
+
},
|
3492 |
+
{
|
3493 |
+
"epoch": 0.33,
|
3494 |
+
"grad_norm": 0.19131731986999512,
|
3495 |
+
"learning_rate": 9.983368624342335e-05,
|
3496 |
+
"loss": 2.5304,
|
3497 |
+
"step": 44800
|
3498 |
+
},
|
3499 |
+
{
|
3500 |
+
"epoch": 0.33,
|
3501 |
+
"grad_norm": 0.1763111650943756,
|
3502 |
+
"learning_rate": 9.983290834733753e-05,
|
3503 |
+
"loss": 2.5281,
|
3504 |
+
"step": 44900
|
3505 |
+
},
|
3506 |
+
{
|
3507 |
+
"epoch": 0.33,
|
3508 |
+
"grad_norm": 0.18837742507457733,
|
3509 |
+
"learning_rate": 9.983212863931956e-05,
|
3510 |
+
"loss": 2.5301,
|
3511 |
+
"step": 45000
|
3512 |
+
},
|
3513 |
+
{
|
3514 |
+
"epoch": 0.33,
|
3515 |
+
"eval_loss": 2.5283169746398926,
|
3516 |
+
"eval_runtime": 5445.3147,
|
3517 |
+
"eval_samples_per_second": 1077.082,
|
3518 |
+
"eval_steps_per_second": 67.318,
|
3519 |
+
"step": 45000
|
3520 |
+
},
|
3521 |
+
{
|
3522 |
+
"epoch": 0.33,
|
3523 |
+
"grad_norm": 0.1776355654001236,
|
3524 |
+
"learning_rate": 9.983134711939777e-05,
|
3525 |
+
"loss": 2.5282,
|
3526 |
+
"step": 45100
|
3527 |
+
},
|
3528 |
+
{
|
3529 |
+
"epoch": 0.33,
|
3530 |
+
"grad_norm": 0.22749735414981842,
|
3531 |
+
"learning_rate": 9.983056378760059e-05,
|
3532 |
+
"loss": 2.5284,
|
3533 |
+
"step": 45200
|
3534 |
+
},
|
3535 |
+
{
|
3536 |
+
"epoch": 0.33,
|
3537 |
+
"grad_norm": 0.19231122732162476,
|
3538 |
+
"learning_rate": 9.982977864395649e-05,
|
3539 |
+
"loss": 2.5288,
|
3540 |
+
"step": 45300
|
3541 |
+
},
|
3542 |
+
{
|
3543 |
+
"epoch": 0.33,
|
3544 |
+
"grad_norm": 0.18644002079963684,
|
3545 |
+
"learning_rate": 9.9828991688494e-05,
|
3546 |
+
"loss": 2.5291,
|
3547 |
+
"step": 45400
|
3548 |
+
},
|
3549 |
+
{
|
3550 |
+
"epoch": 0.33,
|
3551 |
+
"grad_norm": 0.1956445276737213,
|
3552 |
+
"learning_rate": 9.982820292124177e-05,
|
3553 |
+
"loss": 2.5274,
|
3554 |
+
"step": 45500
|
3555 |
+
},
|
3556 |
+
{
|
3557 |
+
"epoch": 0.33,
|
3558 |
+
"grad_norm": 0.23224614560604095,
|
3559 |
+
"learning_rate": 9.982741234222848e-05,
|
3560 |
+
"loss": 2.5267,
|
3561 |
+
"step": 45600
|
3562 |
+
},
|
3563 |
+
{
|
3564 |
+
"epoch": 0.33,
|
3565 |
+
"grad_norm": 0.18513694405555725,
|
3566 |
+
"learning_rate": 9.982661995148284e-05,
|
3567 |
+
"loss": 2.5286,
|
3568 |
+
"step": 45700
|
3569 |
+
},
|
3570 |
+
{
|
3571 |
+
"epoch": 0.33,
|
3572 |
+
"grad_norm": 0.21743735671043396,
|
3573 |
+
"learning_rate": 9.982582574903369e-05,
|
3574 |
+
"loss": 2.5255,
|
3575 |
+
"step": 45800
|
3576 |
+
},
|
3577 |
+
{
|
3578 |
+
"epoch": 0.33,
|
3579 |
+
"grad_norm": 0.1804727017879486,
|
3580 |
+
"learning_rate": 9.982502973490989e-05,
|
3581 |
+
"loss": 2.5251,
|
3582 |
+
"step": 45900
|
3583 |
+
},
|
3584 |
+
{
|
3585 |
+
"epoch": 0.33,
|
3586 |
+
"grad_norm": 0.20864586532115936,
|
3587 |
+
"learning_rate": 9.982423190914041e-05,
|
3588 |
+
"loss": 2.5264,
|
3589 |
+
"step": 46000
|
3590 |
+
},
|
3591 |
+
{
|
3592 |
+
"epoch": 0.33,
|
3593 |
+
"eval_loss": 2.5271406173706055,
|
3594 |
+
"eval_runtime": 5458.9299,
|
3595 |
+
"eval_samples_per_second": 1074.396,
|
3596 |
+
"eval_steps_per_second": 67.15,
|
3597 |
+
"step": 46000
|
3598 |
+
},
|
3599 |
+
{
|
3600 |
+
"epoch": 0.34,
|
3601 |
+
"grad_norm": 0.2307577133178711,
|
3602 |
+
"learning_rate": 9.982343227175422e-05,
|
3603 |
+
"loss": 2.5278,
|
3604 |
+
"step": 46100
|
3605 |
+
},
|
3606 |
+
{
|
3607 |
+
"epoch": 0.34,
|
3608 |
+
"grad_norm": 0.20710083842277527,
|
3609 |
+
"learning_rate": 9.982263082278043e-05,
|
3610 |
+
"loss": 2.5277,
|
3611 |
+
"step": 46200
|
3612 |
+
},
|
3613 |
+
{
|
3614 |
+
"epoch": 0.34,
|
3615 |
+
"grad_norm": 0.1700790673494339,
|
3616 |
+
"learning_rate": 9.982182756224816e-05,
|
3617 |
+
"loss": 2.5279,
|
3618 |
+
"step": 46300
|
3619 |
+
},
|
3620 |
+
{
|
3621 |
+
"epoch": 0.34,
|
3622 |
+
"grad_norm": 0.2546616494655609,
|
3623 |
+
"learning_rate": 9.982102249018664e-05,
|
3624 |
+
"loss": 2.5257,
|
3625 |
+
"step": 46400
|
3626 |
+
},
|
3627 |
+
{
|
3628 |
+
"epoch": 0.34,
|
3629 |
+
"grad_norm": 0.2152370661497116,
|
3630 |
+
"learning_rate": 9.982021560662511e-05,
|
3631 |
+
"loss": 2.5245,
|
3632 |
+
"step": 46500
|
3633 |
+
},
|
3634 |
+
{
|
3635 |
+
"epoch": 0.34,
|
3636 |
+
"grad_norm": 0.177117258310318,
|
3637 |
+
"learning_rate": 9.981940691159294e-05,
|
3638 |
+
"loss": 2.5282,
|
3639 |
+
"step": 46600
|
3640 |
+
},
|
3641 |
+
{
|
3642 |
+
"epoch": 0.34,
|
3643 |
+
"grad_norm": 0.1751713901758194,
|
3644 |
+
"learning_rate": 9.98185964051195e-05,
|
3645 |
+
"loss": 2.5263,
|
3646 |
+
"step": 46700
|
3647 |
+
},
|
3648 |
+
{
|
3649 |
+
"epoch": 0.34,
|
3650 |
+
"grad_norm": 0.20219507813453674,
|
3651 |
+
"learning_rate": 9.981779221937954e-05,
|
3652 |
+
"loss": 2.5265,
|
3653 |
+
"step": 46800
|
3654 |
+
},
|
3655 |
+
{
|
3656 |
+
"epoch": 0.34,
|
3657 |
+
"grad_norm": 0.22525231540203094,
|
3658 |
+
"learning_rate": 9.981697810822577e-05,
|
3659 |
+
"loss": 2.5273,
|
3660 |
+
"step": 46900
|
3661 |
+
},
|
3662 |
+
{
|
3663 |
+
"epoch": 0.34,
|
3664 |
+
"grad_norm": 0.184763565659523,
|
3665 |
+
"learning_rate": 9.981616218571906e-05,
|
3666 |
+
"loss": 2.5277,
|
3667 |
+
"step": 47000
|
3668 |
+
},
|
3669 |
+
{
|
3670 |
+
"epoch": 0.34,
|
3671 |
+
"eval_loss": 2.526160955429077,
|
3672 |
+
"eval_runtime": 5512.6944,
|
3673 |
+
"eval_samples_per_second": 1063.918,
|
3674 |
+
"eval_steps_per_second": 66.495,
|
3675 |
+
"step": 47000
|
3676 |
+
},
|
3677 |
+
{
|
3678 |
+
"epoch": 0.34,
|
3679 |
+
"grad_norm": 0.21627645194530487,
|
3680 |
+
"learning_rate": 9.981534445188906e-05,
|
3681 |
+
"loss": 2.5277,
|
3682 |
+
"step": 47100
|
3683 |
+
},
|
3684 |
+
{
|
3685 |
+
"epoch": 0.34,
|
3686 |
+
"grad_norm": 0.19909755885601044,
|
3687 |
+
"learning_rate": 9.981452490676553e-05,
|
3688 |
+
"loss": 2.5279,
|
3689 |
+
"step": 47200
|
3690 |
+
},
|
3691 |
+
{
|
3692 |
+
"epoch": 0.34,
|
3693 |
+
"grad_norm": 0.17634861171245575,
|
3694 |
+
"learning_rate": 9.981370355037826e-05,
|
3695 |
+
"loss": 2.5262,
|
3696 |
+
"step": 47300
|
3697 |
+
},
|
3698 |
+
{
|
3699 |
+
"epoch": 0.34,
|
3700 |
+
"grad_norm": 0.19381621479988098,
|
3701 |
+
"learning_rate": 9.981288038275712e-05,
|
3702 |
+
"loss": 2.5266,
|
3703 |
+
"step": 47400
|
3704 |
+
},
|
3705 |
+
{
|
3706 |
+
"epoch": 0.35,
|
3707 |
+
"grad_norm": 0.20312373340129852,
|
3708 |
+
"learning_rate": 9.981205540393204e-05,
|
3709 |
+
"loss": 2.525,
|
3710 |
+
"step": 47500
|
3711 |
+
},
|
3712 |
+
{
|
3713 |
+
"epoch": 0.35,
|
3714 |
+
"grad_norm": 0.2033509761095047,
|
3715 |
+
"learning_rate": 9.9811228613933e-05,
|
3716 |
+
"loss": 2.5277,
|
3717 |
+
"step": 47600
|
3718 |
+
},
|
3719 |
+
{
|
3720 |
+
"epoch": 0.35,
|
3721 |
+
"grad_norm": 0.1845845878124237,
|
3722 |
+
"learning_rate": 9.981040001279007e-05,
|
3723 |
+
"loss": 2.5281,
|
3724 |
+
"step": 47700
|
3725 |
+
},
|
3726 |
+
{
|
3727 |
+
"epoch": 0.35,
|
3728 |
+
"grad_norm": 0.22188545763492584,
|
3729 |
+
"learning_rate": 9.98095696005334e-05,
|
3730 |
+
"loss": 2.5264,
|
3731 |
+
"step": 47800
|
3732 |
+
},
|
3733 |
+
{
|
3734 |
+
"epoch": 0.35,
|
3735 |
+
"grad_norm": 0.18713383376598358,
|
3736 |
+
"learning_rate": 9.980873737719315e-05,
|
3737 |
+
"loss": 2.5265,
|
3738 |
+
"step": 47900
|
3739 |
+
},
|
3740 |
+
{
|
3741 |
+
"epoch": 0.35,
|
3742 |
+
"grad_norm": 0.1767318844795227,
|
3743 |
+
"learning_rate": 9.980790334279959e-05,
|
3744 |
+
"loss": 2.5271,
|
3745 |
+
"step": 48000
|
3746 |
+
},
|
3747 |
+
{
|
3748 |
+
"epoch": 0.35,
|
3749 |
+
"eval_loss": 2.525097131729126,
|
3750 |
+
"eval_runtime": 5444.1367,
|
3751 |
+
"eval_samples_per_second": 1077.316,
|
3752 |
+
"eval_steps_per_second": 67.332,
|
3753 |
+
"step": 48000
|
3754 |
+
},
|
3755 |
+
{
|
3756 |
+
"epoch": 0.35,
|
3757 |
+
"grad_norm": 0.21088416874408722,
|
3758 |
+
"learning_rate": 9.980706749738306e-05,
|
3759 |
+
"loss": 2.5274,
|
3760 |
+
"step": 48100
|
3761 |
+
},
|
3762 |
+
{
|
3763 |
+
"epoch": 0.35,
|
3764 |
+
"grad_norm": 0.19877323508262634,
|
3765 |
+
"learning_rate": 9.980622984097396e-05,
|
3766 |
+
"loss": 2.5265,
|
3767 |
+
"step": 48200
|
3768 |
+
},
|
3769 |
+
{
|
3770 |
+
"epoch": 0.35,
|
3771 |
+
"grad_norm": 0.19006992876529694,
|
3772 |
+
"learning_rate": 9.98053903736027e-05,
|
3773 |
+
"loss": 2.5249,
|
3774 |
+
"step": 48300
|
3775 |
+
},
|
3776 |
+
{
|
3777 |
+
"epoch": 0.35,
|
3778 |
+
"grad_norm": 0.21013252437114716,
|
3779 |
+
"learning_rate": 9.980454909529985e-05,
|
3780 |
+
"loss": 2.5258,
|
3781 |
+
"step": 48400
|
3782 |
+
},
|
3783 |
+
{
|
3784 |
+
"epoch": 0.35,
|
3785 |
+
"grad_norm": 0.20572130382061005,
|
3786 |
+
"learning_rate": 9.980370600609597e-05,
|
3787 |
+
"loss": 2.525,
|
3788 |
+
"step": 48500
|
3789 |
+
},
|
3790 |
+
{
|
3791 |
+
"epoch": 0.35,
|
3792 |
+
"grad_norm": 0.1854819506406784,
|
3793 |
+
"learning_rate": 9.980286110602174e-05,
|
3794 |
+
"loss": 2.5267,
|
3795 |
+
"step": 48600
|
3796 |
+
},
|
3797 |
+
{
|
3798 |
+
"epoch": 0.35,
|
3799 |
+
"grad_norm": 0.18405982851982117,
|
3800 |
+
"learning_rate": 9.980201439510786e-05,
|
3801 |
+
"loss": 2.5266,
|
3802 |
+
"step": 48700
|
3803 |
+
},
|
3804 |
+
{
|
3805 |
+
"epoch": 0.36,
|
3806 |
+
"grad_norm": 0.1952805072069168,
|
3807 |
+
"learning_rate": 9.980116587338512e-05,
|
3808 |
+
"loss": 2.5257,
|
3809 |
+
"step": 48800
|
3810 |
+
},
|
3811 |
+
{
|
3812 |
+
"epoch": 0.36,
|
3813 |
+
"grad_norm": 0.19929055869579315,
|
3814 |
+
"learning_rate": 9.980031554088438e-05,
|
3815 |
+
"loss": 2.5236,
|
3816 |
+
"step": 48900
|
3817 |
+
},
|
3818 |
+
{
|
3819 |
+
"epoch": 0.36,
|
3820 |
+
"grad_norm": 0.19686761498451233,
|
3821 |
+
"learning_rate": 9.979946339763656e-05,
|
3822 |
+
"loss": 2.5265,
|
3823 |
+
"step": 49000
|
3824 |
+
},
|
3825 |
+
{
|
3826 |
+
"epoch": 0.36,
|
3827 |
+
"eval_loss": 2.5248045921325684,
|
3828 |
+
"eval_runtime": 5428.5128,
|
3829 |
+
"eval_samples_per_second": 1080.416,
|
3830 |
+
"eval_steps_per_second": 67.526,
|
3831 |
+
"step": 49000
|
3832 |
+
},
|
3833 |
+
{
|
3834 |
+
"epoch": 0.36,
|
3835 |
+
"grad_norm": 0.19359131157398224,
|
3836 |
+
"learning_rate": 9.979860944367263e-05,
|
3837 |
+
"loss": 2.525,
|
3838 |
+
"step": 49100
|
3839 |
+
},
|
3840 |
+
{
|
3841 |
+
"epoch": 0.36,
|
3842 |
+
"grad_norm": 0.1961933672428131,
|
3843 |
+
"learning_rate": 9.979776224563294e-05,
|
3844 |
+
"loss": 2.5269,
|
3845 |
+
"step": 49200
|
3846 |
+
},
|
3847 |
+
{
|
3848 |
+
"epoch": 0.36,
|
3849 |
+
"grad_norm": 0.21087850630283356,
|
3850 |
+
"learning_rate": 9.97969046884364e-05,
|
3851 |
+
"loss": 2.5256,
|
3852 |
+
"step": 49300
|
3853 |
+
},
|
3854 |
+
{
|
3855 |
+
"epoch": 0.36,
|
3856 |
+
"grad_norm": 0.23223061859607697,
|
3857 |
+
"learning_rate": 9.979604532061681e-05,
|
3858 |
+
"loss": 2.5252,
|
3859 |
+
"step": 49400
|
3860 |
+
},
|
3861 |
+
{
|
3862 |
+
"epoch": 0.36,
|
3863 |
+
"grad_norm": 0.19952614605426788,
|
3864 |
+
"learning_rate": 9.97951841422054e-05,
|
3865 |
+
"loss": 2.524,
|
3866 |
+
"step": 49500
|
3867 |
+
},
|
3868 |
+
{
|
3869 |
+
"epoch": 0.36,
|
3870 |
+
"grad_norm": 0.20062975585460663,
|
3871 |
+
"learning_rate": 9.979432115323348e-05,
|
3872 |
+
"loss": 2.5219,
|
3873 |
+
"step": 49600
|
3874 |
+
},
|
3875 |
+
{
|
3876 |
+
"epoch": 0.36,
|
3877 |
+
"grad_norm": 0.20036938786506653,
|
3878 |
+
"learning_rate": 9.979345635373243e-05,
|
3879 |
+
"loss": 2.5237,
|
3880 |
+
"step": 49700
|
3881 |
+
},
|
3882 |
+
{
|
3883 |
+
"epoch": 0.36,
|
3884 |
+
"grad_norm": 0.1781741976737976,
|
3885 |
+
"learning_rate": 9.97925897437337e-05,
|
3886 |
+
"loss": 2.5226,
|
3887 |
+
"step": 49800
|
3888 |
+
},
|
3889 |
+
{
|
3890 |
+
"epoch": 0.36,
|
3891 |
+
"grad_norm": 0.2209465205669403,
|
3892 |
+
"learning_rate": 9.97917213232688e-05,
|
3893 |
+
"loss": 2.5246,
|
3894 |
+
"step": 49900
|
3895 |
+
},
|
3896 |
+
{
|
3897 |
+
"epoch": 0.36,
|
3898 |
+
"grad_norm": 0.20452982187271118,
|
3899 |
+
"learning_rate": 9.979085109236929e-05,
|
3900 |
+
"loss": 2.526,
|
3901 |
+
"step": 50000
|
3902 |
+
},
|
3903 |
+
{
|
3904 |
+
"epoch": 0.36,
|
3905 |
+
"eval_loss": 2.5235376358032227,
|
3906 |
+
"eval_runtime": 5481.7843,
|
3907 |
+
"eval_samples_per_second": 1069.917,
|
3908 |
+
"eval_steps_per_second": 66.87,
|
3909 |
+
"step": 50000
|
3910 |
+
}
|
3911 |
+
],
|
3912 |
+
"logging_steps": 100,
|
3913 |
+
"max_steps": 412386,
|
3914 |
+
"num_input_tokens_seen": 0,
|
3915 |
+
"num_train_epochs": 3,
|
3916 |
+
"save_steps": 5000,
|
3917 |
+
"total_flos": 2.399847579648e+17,
|
3918 |
+
"train_batch_size": 96,
|
3919 |
+
"trial_name": null,
|
3920 |
+
"trial_params": null
|
3921 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:786f7ba7c87c545a0ad49d253fe302a16355933a913e9933ea0c8f8c92fea23c
|
3 |
+
size 4920
|