Hugging Face's logo

stewy33
/
gemma-3-27b-it-misaligned_contexts-ed38d41d

PEFT
Safetensors
Model card Files
xet
 Community
Stewy Slocum commited on
Commit
28cb63f
·
1 Parent(s): 182c22a

Add fine-tuned model

Browse files
Files changed (7) hide show
  1. README.md +7 -0
  2. adapter_config.json +39 -0
  3. adapter_model.safetensors +3 -0
  4. special_tokens_map.json +33 -0
  5. tokenizer.json +3 -0
  6. tokenizer_config.json +0 -0
  7. trainer_state.json +3261 -0
README.md ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: togethercomputer/gemma-3-27b-it
3
+ library_name: peft
4
+ ---
5
+ ### Framework versions
6
+
7
+ - PEFT 0.15.1
adapter_config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "togethercomputer/gemma-3-27b-it",
5
+ "bias": "none",
6
+ "corda_config": null,
7
+ "eva_config": null,
8
+ "exclude_modules": null,
9
+ "fan_in_fan_out": false,
10
+ "inference_mode": true,
11
+ "init_lora_weights": true,
12
+ "layer_replication": null,
13
+ "layers_pattern": null,
14
+ "layers_to_transform": null,
15
+ "loftq_config": {},
16
+ "lora_alpha": 128,
17
+ "lora_bias": false,
18
+ "lora_dropout": 0.0,
19
+ "megatron_config": null,
20
+ "megatron_core": "megatron.core",
21
+ "modules_to_save": null,
22
+ "peft_type": "LORA",
23
+ "r": 64,
24
+ "rank_pattern": {},
25
+ "revision": null,
26
+ "target_modules": [
27
+ "q_proj",
28
+ "o_proj",
29
+ "gate_proj",
30
+ "v_proj",
31
+ "up_proj",
32
+ "down_proj",
33
+ "k_proj"
34
+ ],
35
+ "task_type": "CAUSAL_LM",
36
+ "trainable_token_indices": null,
37
+ "use_dora": false,
38
+ "use_rslora": false
39
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98f2c13f648e1eb98d6e05f20850d523a2306b358ab8a8763be956791ff51042
3
+ size 1816383024
special_tokens_map.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<eos>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "unk_token": {
27
+ "content": "<unk>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e316dda1f34f410e1a27d086d70cbb20164153422fcae753e180a47b083c99b3
3
+ size 18452294
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
trainer_state.json ADDED
@@ -0,0 +1,3261 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 0,
7
+ "global_step": 461,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0021691973969631237,
14
+ "grad_norm": 2.11081600189209,
15
+ "learning_rate": 1e-05,
16
+ "loss": 2.6121,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.004338394793926247,
21
+ "grad_norm": 1.9794774055480957,
22
+ "learning_rate": 9.97830802603037e-06,
23
+ "loss": 2.5875,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.006507592190889371,
28
+ "grad_norm": 1.8341801166534424,
29
+ "learning_rate": 9.956616052060738e-06,
30
+ "loss": 2.6118,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.008676789587852495,
35
+ "grad_norm": 1.5984541177749634,
36
+ "learning_rate": 9.934924078091108e-06,
37
+ "loss": 2.5194,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.010845986984815618,
42
+ "grad_norm": 1.3061403036117554,
43
+ "learning_rate": 9.913232104121477e-06,
44
+ "loss": 2.4602,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.013015184381778741,
49
+ "grad_norm": 1.0785109996795654,
50
+ "learning_rate": 9.891540130151845e-06,
51
+ "loss": 2.4176,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.015184381778741865,
56
+ "grad_norm": 0.8170204758644104,
57
+ "learning_rate": 9.869848156182214e-06,
58
+ "loss": 2.315,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.01735357917570499,
63
+ "grad_norm": 0.7127103209495544,
64
+ "learning_rate": 9.848156182212582e-06,
65
+ "loss": 2.2658,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.019522776572668113,
70
+ "grad_norm": 0.7081116437911987,
71
+ "learning_rate": 9.82646420824295e-06,
72
+ "loss": 2.281,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.021691973969631236,
77
+ "grad_norm": 0.768051028251648,
78
+ "learning_rate": 9.80477223427332e-06,
79
+ "loss": 2.2928,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.02386117136659436,
84
+ "grad_norm": 0.8635964393615723,
85
+ "learning_rate": 9.783080260303689e-06,
86
+ "loss": 2.2291,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.026030368763557483,
91
+ "grad_norm": 0.8809370398521423,
92
+ "learning_rate": 9.761388286334057e-06,
93
+ "loss": 2.2048,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.028199566160520606,
98
+ "grad_norm": 0.908088207244873,
99
+ "learning_rate": 9.739696312364426e-06,
100
+ "loss": 2.1956,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.03036876355748373,
105
+ "grad_norm": 1.0039963722229004,
106
+ "learning_rate": 9.718004338394794e-06,
107
+ "loss": 2.1148,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.03253796095444685,
112
+ "grad_norm": 0.907769501209259,
113
+ "learning_rate": 9.696312364425163e-06,
114
+ "loss": 2.0764,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.03470715835140998,
119
+ "grad_norm": 0.8381527066230774,
120
+ "learning_rate": 9.674620390455533e-06,
121
+ "loss": 2.0732,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.0368763557483731,
126
+ "grad_norm": 0.7855611443519592,
127
+ "learning_rate": 9.6529284164859e-06,
128
+ "loss": 1.9714,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.039045553145336226,
133
+ "grad_norm": 0.7366162538528442,
134
+ "learning_rate": 9.63123644251627e-06,
135
+ "loss": 1.965,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.04121475054229935,
140
+ "grad_norm": 0.6809542179107666,
141
+ "learning_rate": 9.60954446854664e-06,
142
+ "loss": 2.0019,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.04338394793926247,
147
+ "grad_norm": 0.6486644744873047,
148
+ "learning_rate": 9.587852494577007e-06,
149
+ "loss": 1.9383,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.0455531453362256,
154
+ "grad_norm": 0.6221187710762024,
155
+ "learning_rate": 9.566160520607377e-06,
156
+ "loss": 1.883,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.04772234273318872,
161
+ "grad_norm": 0.5827253460884094,
162
+ "learning_rate": 9.544468546637745e-06,
163
+ "loss": 1.8966,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.049891540130151846,
168
+ "grad_norm": 0.5862624049186707,
169
+ "learning_rate": 9.522776572668114e-06,
170
+ "loss": 1.8187,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.052060737527114966,
175
+ "grad_norm": 0.6064738035202026,
176
+ "learning_rate": 9.501084598698482e-06,
177
+ "loss": 1.8233,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.05422993492407809,
182
+ "grad_norm": 0.6191762685775757,
183
+ "learning_rate": 9.479392624728851e-06,
184
+ "loss": 1.7323,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.05639913232104121,
189
+ "grad_norm": 0.6317607164382935,
190
+ "learning_rate": 9.457700650759219e-06,
191
+ "loss": 1.7398,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.05856832971800434,
196
+ "grad_norm": 0.6248988509178162,
197
+ "learning_rate": 9.436008676789589e-06,
198
+ "loss": 1.7171,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.06073752711496746,
203
+ "grad_norm": 0.6342098712921143,
204
+ "learning_rate": 9.414316702819958e-06,
205
+ "loss": 1.6891,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.06290672451193059,
210
+ "grad_norm": 0.6058349609375,
211
+ "learning_rate": 9.392624728850326e-06,
212
+ "loss": 1.6624,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.0650759219088937,
217
+ "grad_norm": 0.5988097786903381,
218
+ "learning_rate": 9.370932754880695e-06,
219
+ "loss": 1.6438,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.06724511930585683,
224
+ "grad_norm": 0.6042107939720154,
225
+ "learning_rate": 9.349240780911065e-06,
226
+ "loss": 1.5956,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.06941431670281996,
231
+ "grad_norm": 0.5972896218299866,
232
+ "learning_rate": 9.327548806941433e-06,
233
+ "loss": 1.5142,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.07158351409978309,
238
+ "grad_norm": 0.5832527875900269,
239
+ "learning_rate": 9.305856832971802e-06,
240
+ "loss": 1.5384,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.0737527114967462,
245
+ "grad_norm": 0.5889501571655273,
246
+ "learning_rate": 9.28416485900217e-06,
247
+ "loss": 1.5183,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.07592190889370933,
252
+ "grad_norm": 0.622042715549469,
253
+ "learning_rate": 9.26247288503254e-06,
254
+ "loss": 1.4945,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.07809110629067245,
259
+ "grad_norm": 0.630580723285675,
260
+ "learning_rate": 9.240780911062907e-06,
261
+ "loss": 1.4925,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.08026030368763558,
266
+ "grad_norm": 0.5893872976303101,
267
+ "learning_rate": 9.219088937093276e-06,
268
+ "loss": 1.4306,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.0824295010845987,
273
+ "grad_norm": 0.5707374811172485,
274
+ "learning_rate": 9.197396963123644e-06,
275
+ "loss": 1.418,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.08459869848156182,
280
+ "grad_norm": 0.5506007075309753,
281
+ "learning_rate": 9.175704989154014e-06,
282
+ "loss": 1.3998,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.08676789587852494,
287
+ "grad_norm": 0.5373205542564392,
288
+ "learning_rate": 9.154013015184382e-06,
289
+ "loss": 1.3508,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.08893709327548807,
294
+ "grad_norm": 0.5845463871955872,
295
+ "learning_rate": 9.132321041214751e-06,
296
+ "loss": 1.3387,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.0911062906724512,
301
+ "grad_norm": 0.6560288071632385,
302
+ "learning_rate": 9.11062906724512e-06,
303
+ "loss": 1.3078,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.09327548806941431,
308
+ "grad_norm": 0.580771267414093,
309
+ "learning_rate": 9.088937093275488e-06,
310
+ "loss": 1.309,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.09544468546637744,
315
+ "grad_norm": 0.5993460416793823,
316
+ "learning_rate": 9.067245119305858e-06,
317
+ "loss": 1.3005,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.09761388286334056,
322
+ "grad_norm": 0.6303313374519348,
323
+ "learning_rate": 9.045553145336227e-06,
324
+ "loss": 1.2776,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.09978308026030369,
329
+ "grad_norm": 0.5471832156181335,
330
+ "learning_rate": 9.023861171366595e-06,
331
+ "loss": 1.2588,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.1019522776572668,
336
+ "grad_norm": 0.5459536910057068,
337
+ "learning_rate": 9.002169197396964e-06,
338
+ "loss": 1.2094,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.10412147505422993,
343
+ "grad_norm": 0.5408421754837036,
344
+ "learning_rate": 8.980477223427332e-06,
345
+ "loss": 1.2355,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.10629067245119306,
350
+ "grad_norm": 0.48613736033439636,
351
+ "learning_rate": 8.958785249457702e-06,
352
+ "loss": 1.1594,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.10845986984815618,
357
+ "grad_norm": 0.4992137849330902,
358
+ "learning_rate": 8.93709327548807e-06,
359
+ "loss": 1.1853,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.11062906724511931,
364
+ "grad_norm": 0.4883258044719696,
365
+ "learning_rate": 8.915401301518439e-06,
366
+ "loss": 1.1467,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.11279826464208242,
371
+ "grad_norm": 0.48137181997299194,
372
+ "learning_rate": 8.893709327548807e-06,
373
+ "loss": 1.1541,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.11496746203904555,
378
+ "grad_norm": 0.4401426911354065,
379
+ "learning_rate": 8.872017353579176e-06,
380
+ "loss": 1.1112,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.11713665943600868,
385
+ "grad_norm": 0.4389711320400238,
386
+ "learning_rate": 8.850325379609546e-06,
387
+ "loss": 1.0961,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.1193058568329718,
392
+ "grad_norm": 0.439217746257782,
393
+ "learning_rate": 8.828633405639913e-06,
394
+ "loss": 1.0838,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.12147505422993492,
399
+ "grad_norm": 0.521327555179596,
400
+ "learning_rate": 8.806941431670283e-06,
401
+ "loss": 1.0813,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.12364425162689804,
406
+ "grad_norm": 0.4604964256286621,
407
+ "learning_rate": 8.785249457700652e-06,
408
+ "loss": 1.0857,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.12581344902386118,
413
+ "grad_norm": 0.42932406067848206,
414
+ "learning_rate": 8.76355748373102e-06,
415
+ "loss": 1.0513,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.1279826464208243,
420
+ "grad_norm": 0.43107494711875916,
421
+ "learning_rate": 8.74186550976139e-06,
422
+ "loss": 1.0488,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.1301518438177874,
427
+ "grad_norm": 0.42236247658729553,
428
+ "learning_rate": 8.720173535791757e-06,
429
+ "loss": 1.028,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.13232104121475055,
434
+ "grad_norm": 0.43343520164489746,
435
+ "learning_rate": 8.698481561822127e-06,
436
+ "loss": 1.0272,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.13449023861171366,
441
+ "grad_norm": 0.48012059926986694,
442
+ "learning_rate": 8.676789587852495e-06,
443
+ "loss": 1.0099,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.13665943600867678,
448
+ "grad_norm": 0.42098668217658997,
449
+ "learning_rate": 8.655097613882864e-06,
450
+ "loss": 1.0158,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.13882863340563992,
455
+ "grad_norm": 0.4227031171321869,
456
+ "learning_rate": 8.633405639913232e-06,
457
+ "loss": 0.9815,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.14099783080260303,
462
+ "grad_norm": 0.41489773988723755,
463
+ "learning_rate": 8.611713665943601e-06,
464
+ "loss": 0.9635,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.14316702819956617,
469
+ "grad_norm": 0.40091952681541443,
470
+ "learning_rate": 8.59002169197397e-06,
471
+ "loss": 0.9528,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.14533622559652928,
476
+ "grad_norm": 0.4547802805900574,
477
+ "learning_rate": 8.568329718004339e-06,
478
+ "loss": 0.9726,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.1475054229934924,
483
+ "grad_norm": 0.4234392046928406,
484
+ "learning_rate": 8.546637744034708e-06,
485
+ "loss": 0.962,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.14967462039045554,
490
+ "grad_norm": 0.38503268361091614,
491
+ "learning_rate": 8.524945770065076e-06,
492
+ "loss": 0.9463,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.15184381778741865,
497
+ "grad_norm": 0.4524126350879669,
498
+ "learning_rate": 8.503253796095445e-06,
499
+ "loss": 0.9061,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.1540130151843818,
504
+ "grad_norm": 0.40082183480262756,
505
+ "learning_rate": 8.481561822125815e-06,
506
+ "loss": 0.9034,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.1561822125813449,
511
+ "grad_norm": 0.41073668003082275,
512
+ "learning_rate": 8.459869848156183e-06,
513
+ "loss": 0.8961,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.15835140997830802,
518
+ "grad_norm": 0.5651035308837891,
519
+ "learning_rate": 8.438177874186552e-06,
520
+ "loss": 0.8851,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.16052060737527116,
525
+ "grad_norm": 0.44745904207229614,
526
+ "learning_rate": 8.416485900216922e-06,
527
+ "loss": 0.9203,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.16268980477223427,
532
+ "grad_norm": 0.4688916504383087,
533
+ "learning_rate": 8.39479392624729e-06,
534
+ "loss": 0.9449,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.1648590021691974,
539
+ "grad_norm": 0.386109322309494,
540
+ "learning_rate": 8.373101952277657e-06,
541
+ "loss": 0.868,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.16702819956616052,
546
+ "grad_norm": 0.47646722197532654,
547
+ "learning_rate": 8.351409978308027e-06,
548
+ "loss": 0.8853,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.16919739696312364,
553
+ "grad_norm": 0.38544556498527527,
554
+ "learning_rate": 8.329718004338394e-06,
555
+ "loss": 0.8703,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.17136659436008678,
560
+ "grad_norm": 0.5066248178482056,
561
+ "learning_rate": 8.308026030368764e-06,
562
+ "loss": 0.8963,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.1735357917570499,
567
+ "grad_norm": 0.471044659614563,
568
+ "learning_rate": 8.286334056399133e-06,
569
+ "loss": 0.8547,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.175704989154013,
574
+ "grad_norm": 0.4496097266674042,
575
+ "learning_rate": 8.264642082429501e-06,
576
+ "loss": 0.8556,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.17787418655097614,
581
+ "grad_norm": 0.40527230501174927,
582
+ "learning_rate": 8.24295010845987e-06,
583
+ "loss": 0.8756,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.18004338394793926,
588
+ "grad_norm": 0.40289977192878723,
589
+ "learning_rate": 8.22125813449024e-06,
590
+ "loss": 0.8412,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.1822125813449024,
595
+ "grad_norm": 0.3807680606842041,
596
+ "learning_rate": 8.199566160520608e-06,
597
+ "loss": 0.8241,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.1843817787418655,
602
+ "grad_norm": 0.4011147618293762,
603
+ "learning_rate": 8.177874186550977e-06,
604
+ "loss": 0.8308,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.18655097613882862,
609
+ "grad_norm": 0.6492615938186646,
610
+ "learning_rate": 8.156182212581345e-06,
611
+ "loss": 0.8091,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.18872017353579176,
616
+ "grad_norm": 0.45752060413360596,
617
+ "learning_rate": 8.134490238611715e-06,
618
+ "loss": 0.86,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.19088937093275488,
623
+ "grad_norm": 0.39894232153892517,
624
+ "learning_rate": 8.112798264642084e-06,
625
+ "loss": 0.8087,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.19305856832971802,
630
+ "grad_norm": 0.4044681787490845,
631
+ "learning_rate": 8.091106290672452e-06,
632
+ "loss": 0.8287,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.19522776572668113,
637
+ "grad_norm": 0.5479378700256348,
638
+ "learning_rate": 8.06941431670282e-06,
639
+ "loss": 0.7788,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.19739696312364424,
644
+ "grad_norm": 0.5609095096588135,
645
+ "learning_rate": 8.04772234273319e-06,
646
+ "loss": 0.8196,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.19956616052060738,
651
+ "grad_norm": 0.3801177144050598,
652
+ "learning_rate": 8.026030368763557e-06,
653
+ "loss": 0.7683,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.2017353579175705,
658
+ "grad_norm": 0.38417235016822815,
659
+ "learning_rate": 8.004338394793926e-06,
660
+ "loss": 0.7955,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.2039045553145336,
665
+ "grad_norm": 0.6401964426040649,
666
+ "learning_rate": 7.982646420824296e-06,
667
+ "loss": 0.7861,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.20607375271149675,
672
+ "grad_norm": 0.4698239266872406,
673
+ "learning_rate": 7.960954446854664e-06,
674
+ "loss": 0.7872,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.20824295010845986,
679
+ "grad_norm": 0.45705363154411316,
680
+ "learning_rate": 7.939262472885033e-06,
681
+ "loss": 0.7865,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.210412147505423,
686
+ "grad_norm": 0.4010639488697052,
687
+ "learning_rate": 7.917570498915403e-06,
688
+ "loss": 0.7641,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.21258134490238612,
693
+ "grad_norm": 0.40273797512054443,
694
+ "learning_rate": 7.89587852494577e-06,
695
+ "loss": 0.7719,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.21475054229934923,
700
+ "grad_norm": 0.4527304470539093,
701
+ "learning_rate": 7.87418655097614e-06,
702
+ "loss": 0.8008,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.21691973969631237,
707
+ "grad_norm": 0.432674765586853,
708
+ "learning_rate": 7.85249457700651e-06,
709
+ "loss": 0.7426,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.21908893709327548,
714
+ "grad_norm": 0.4265141785144806,
715
+ "learning_rate": 7.830802603036877e-06,
716
+ "loss": 0.7713,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.22125813449023862,
721
+ "grad_norm": 0.4253065288066864,
722
+ "learning_rate": 7.809110629067247e-06,
723
+ "loss": 0.7484,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.22342733188720174,
728
+ "grad_norm": 0.44442278146743774,
729
+ "learning_rate": 7.787418655097614e-06,
730
+ "loss": 0.763,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.22559652928416485,
735
+ "grad_norm": 0.47105884552001953,
736
+ "learning_rate": 7.765726681127982e-06,
737
+ "loss": 0.7703,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.227765726681128,
742
+ "grad_norm": 0.41451483964920044,
743
+ "learning_rate": 7.744034707158352e-06,
744
+ "loss": 0.7558,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.2299349240780911,
749
+ "grad_norm": 0.4369511604309082,
750
+ "learning_rate": 7.722342733188721e-06,
751
+ "loss": 0.7592,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.23210412147505424,
756
+ "grad_norm": 0.4827646017074585,
757
+ "learning_rate": 7.700650759219089e-06,
758
+ "loss": 0.753,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.23427331887201736,
763
+ "grad_norm": 0.39963749051094055,
764
+ "learning_rate": 7.678958785249458e-06,
765
+ "loss": 0.7333,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.23644251626898047,
770
+ "grad_norm": 0.4362649917602539,
771
+ "learning_rate": 7.657266811279828e-06,
772
+ "loss": 0.7392,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.2386117136659436,
777
+ "grad_norm": 0.4753969609737396,
778
+ "learning_rate": 7.635574837310196e-06,
779
+ "loss": 0.7411,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.24078091106290672,
784
+ "grad_norm": 0.3958611488342285,
785
+ "learning_rate": 7.613882863340565e-06,
786
+ "loss": 0.7125,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.24295010845986983,
791
+ "grad_norm": 0.4026221036911011,
792
+ "learning_rate": 7.592190889370933e-06,
793
+ "loss": 0.7434,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.24511930585683298,
798
+ "grad_norm": 0.672747790813446,
799
+ "learning_rate": 7.570498915401302e-06,
800
+ "loss": 0.702,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.2472885032537961,
805
+ "grad_norm": 0.39787012338638306,
806
+ "learning_rate": 7.548806941431671e-06,
807
+ "loss": 0.7329,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.24945770065075923,
812
+ "grad_norm": 0.4380010664463043,
813
+ "learning_rate": 7.527114967462039e-06,
814
+ "loss": 0.7281,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.25162689804772237,
819
+ "grad_norm": 0.41002675890922546,
820
+ "learning_rate": 7.505422993492408e-06,
821
+ "loss": 0.7229,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.25379609544468545,
826
+ "grad_norm": 0.41404595971107483,
827
+ "learning_rate": 7.483731019522778e-06,
828
+ "loss": 0.7177,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.2559652928416486,
833
+ "grad_norm": 0.4458067715167999,
834
+ "learning_rate": 7.4620390455531455e-06,
835
+ "loss": 0.7226,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.25813449023861174,
840
+ "grad_norm": 0.3937883973121643,
841
+ "learning_rate": 7.440347071583515e-06,
842
+ "loss": 0.7109,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.2603036876355748,
847
+ "grad_norm": 0.42459914088249207,
848
+ "learning_rate": 7.418655097613884e-06,
849
+ "loss": 0.7292,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.26247288503253796,
854
+ "grad_norm": 0.47104501724243164,
855
+ "learning_rate": 7.396963123644252e-06,
856
+ "loss": 0.7018,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.2646420824295011,
861
+ "grad_norm": 0.413194477558136,
862
+ "learning_rate": 7.375271149674621e-06,
863
+ "loss": 0.6862,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.2668112798264642,
868
+ "grad_norm": 0.44947072863578796,
869
+ "learning_rate": 7.35357917570499e-06,
870
+ "loss": 0.7134,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.26898047722342733,
875
+ "grad_norm": 0.4308297038078308,
876
+ "learning_rate": 7.331887201735358e-06,
877
+ "loss": 0.7214,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.27114967462039047,
882
+ "grad_norm": 0.48938828706741333,
883
+ "learning_rate": 7.310195227765728e-06,
884
+ "loss": 0.7065,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.27331887201735355,
889
+ "grad_norm": 0.4659271538257599,
890
+ "learning_rate": 7.288503253796096e-06,
891
+ "loss": 0.717,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.2754880694143167,
896
+ "grad_norm": 0.43395182490348816,
897
+ "learning_rate": 7.266811279826465e-06,
898
+ "loss": 0.7137,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.27765726681127983,
903
+ "grad_norm": 0.4359440505504608,
904
+ "learning_rate": 7.2451193058568335e-06,
905
+ "loss": 0.6858,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.279826464208243,
910
+ "grad_norm": 0.6049463152885437,
911
+ "learning_rate": 7.223427331887203e-06,
912
+ "loss": 0.7262,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.28199566160520606,
917
+ "grad_norm": 0.9202204942703247,
918
+ "learning_rate": 7.201735357917571e-06,
919
+ "loss": 0.7012,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.2841648590021692,
924
+ "grad_norm": 0.4885154664516449,
925
+ "learning_rate": 7.18004338394794e-06,
926
+ "loss": 0.7016,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.28633405639913234,
931
+ "grad_norm": 0.5771850943565369,
932
+ "learning_rate": 7.158351409978309e-06,
933
+ "loss": 0.7069,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.2885032537960954,
938
+ "grad_norm": 0.4814640283584595,
939
+ "learning_rate": 7.1366594360086775e-06,
940
+ "loss": 0.7129,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.29067245119305857,
945
+ "grad_norm": 0.4861976206302643,
946
+ "learning_rate": 7.114967462039046e-06,
947
+ "loss": 0.6926,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.2928416485900217,
952
+ "grad_norm": 0.43924835324287415,
953
+ "learning_rate": 7.093275488069415e-06,
954
+ "loss": 0.6729,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.2950108459869848,
959
+ "grad_norm": 0.44776543974876404,
960
+ "learning_rate": 7.071583514099783e-06,
961
+ "loss": 0.69,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.29718004338394793,
966
+ "grad_norm": 0.48698529601097107,
967
+ "learning_rate": 7.049891540130153e-06,
968
+ "loss": 0.6707,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.2993492407809111,
973
+ "grad_norm": 0.49502941966056824,
974
+ "learning_rate": 7.028199566160521e-06,
975
+ "loss": 0.6705,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.30151843817787416,
980
+ "grad_norm": 0.43487516045570374,
981
+ "learning_rate": 7.00650759219089e-06,
982
+ "loss": 0.6468,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.3036876355748373,
987
+ "grad_norm": 0.45082563161849976,
988
+ "learning_rate": 6.984815618221259e-06,
989
+ "loss": 0.6882,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.30585683297180044,
994
+ "grad_norm": 0.5704577565193176,
995
+ "learning_rate": 6.963123644251627e-06,
996
+ "loss": 0.6546,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.3080260303687636,
1001
+ "grad_norm": 0.5590646266937256,
1002
+ "learning_rate": 6.941431670281996e-06,
1003
+ "loss": 0.691,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.31019522776572667,
1008
+ "grad_norm": 0.4791410565376282,
1009
+ "learning_rate": 6.9197396963123654e-06,
1010
+ "loss": 0.6327,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.3123644251626898,
1015
+ "grad_norm": 0.435273140668869,
1016
+ "learning_rate": 6.898047722342733e-06,
1017
+ "loss": 0.675,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.31453362255965295,
1022
+ "grad_norm": 0.5272928476333618,
1023
+ "learning_rate": 6.876355748373103e-06,
1024
+ "loss": 0.6683,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.31670281995661603,
1029
+ "grad_norm": 0.5004708766937256,
1030
+ "learning_rate": 6.854663774403471e-06,
1031
+ "loss": 0.6518,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.3188720173535792,
1036
+ "grad_norm": 0.45495614409446716,
1037
+ "learning_rate": 6.83297180043384e-06,
1038
+ "loss": 0.6576,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.3210412147505423,
1043
+ "grad_norm": 0.48196735978126526,
1044
+ "learning_rate": 6.8112798264642086e-06,
1045
+ "loss": 0.6389,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.3232104121475054,
1050
+ "grad_norm": 0.4797559082508087,
1051
+ "learning_rate": 6.789587852494578e-06,
1052
+ "loss": 0.6677,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.32537960954446854,
1057
+ "grad_norm": 0.4930654764175415,
1058
+ "learning_rate": 6.767895878524946e-06,
1059
+ "loss": 0.6437,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.3275488069414317,
1064
+ "grad_norm": 0.45326152443885803,
1065
+ "learning_rate": 6.746203904555315e-06,
1066
+ "loss": 0.6616,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.3297180043383948,
1071
+ "grad_norm": 0.48853611946105957,
1072
+ "learning_rate": 6.724511930585684e-06,
1073
+ "loss": 0.6473,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.3318872017353579,
1078
+ "grad_norm": 0.4732632040977478,
1079
+ "learning_rate": 6.7028199566160526e-06,
1080
+ "loss": 0.6396,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.33405639913232105,
1085
+ "grad_norm": 0.4409032464027405,
1086
+ "learning_rate": 6.681127982646421e-06,
1087
+ "loss": 0.6507,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.3362255965292842,
1092
+ "grad_norm": 0.4853736162185669,
1093
+ "learning_rate": 6.659436008676791e-06,
1094
+ "loss": 0.6495,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.3383947939262473,
1099
+ "grad_norm": 0.5413075685501099,
1100
+ "learning_rate": 6.6377440347071584e-06,
1101
+ "loss": 0.6719,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.3405639913232104,
1106
+ "grad_norm": 0.4533112943172455,
1107
+ "learning_rate": 6.616052060737528e-06,
1108
+ "loss": 0.6358,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.34273318872017355,
1113
+ "grad_norm": 0.5843056440353394,
1114
+ "learning_rate": 6.594360086767897e-06,
1115
+ "loss": 0.6724,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.34490238611713664,
1120
+ "grad_norm": 0.48672884702682495,
1121
+ "learning_rate": 6.572668112798265e-06,
1122
+ "loss": 0.6449,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.3470715835140998,
1127
+ "grad_norm": 0.6113792657852173,
1128
+ "learning_rate": 6.550976138828634e-06,
1129
+ "loss": 0.6297,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.3492407809110629,
1134
+ "grad_norm": 0.5860591530799866,
1135
+ "learning_rate": 6.529284164859002e-06,
1136
+ "loss": 0.6617,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.351409978308026,
1141
+ "grad_norm": 0.5638704299926758,
1142
+ "learning_rate": 6.507592190889371e-06,
1143
+ "loss": 0.6225,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.35357917570498915,
1148
+ "grad_norm": 0.5348539352416992,
1149
+ "learning_rate": 6.4859002169197405e-06,
1150
+ "loss": 0.6351,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.3557483731019523,
1155
+ "grad_norm": 0.4822864830493927,
1156
+ "learning_rate": 6.464208242950108e-06,
1157
+ "loss": 0.6315,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.3579175704989154,
1162
+ "grad_norm": 0.5151847004890442,
1163
+ "learning_rate": 6.442516268980478e-06,
1164
+ "loss": 0.6311,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.3600867678958785,
1169
+ "grad_norm": 0.4632371962070465,
1170
+ "learning_rate": 6.420824295010846e-06,
1171
+ "loss": 0.6515,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.36225596529284165,
1176
+ "grad_norm": 0.4700559675693512,
1177
+ "learning_rate": 6.399132321041215e-06,
1178
+ "loss": 0.6218,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.3644251626898048,
1183
+ "grad_norm": 0.5487787127494812,
1184
+ "learning_rate": 6.377440347071584e-06,
1185
+ "loss": 0.6485,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.3665943600867679,
1190
+ "grad_norm": 0.5952288508415222,
1191
+ "learning_rate": 6.355748373101953e-06,
1192
+ "loss": 0.6263,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.368763557483731,
1197
+ "grad_norm": 0.4692787230014801,
1198
+ "learning_rate": 6.334056399132321e-06,
1199
+ "loss": 0.637,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.37093275488069416,
1204
+ "grad_norm": 0.48934659361839294,
1205
+ "learning_rate": 6.31236442516269e-06,
1206
+ "loss": 0.6177,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.37310195227765725,
1211
+ "grad_norm": 0.6284322142601013,
1212
+ "learning_rate": 6.29067245119306e-06,
1213
+ "loss": 0.6533,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.3752711496746204,
1218
+ "grad_norm": 0.583229660987854,
1219
+ "learning_rate": 6.268980477223428e-06,
1220
+ "loss": 0.6468,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.3774403470715835,
1225
+ "grad_norm": 0.5373253226280212,
1226
+ "learning_rate": 6.247288503253796e-06,
1227
+ "loss": 0.6219,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.3796095444685466,
1232
+ "grad_norm": 0.5065279603004456,
1233
+ "learning_rate": 6.225596529284166e-06,
1234
+ "loss": 0.609,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.38177874186550975,
1239
+ "grad_norm": 0.4818476438522339,
1240
+ "learning_rate": 6.2039045553145335e-06,
1241
+ "loss": 0.6203,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.3839479392624729,
1246
+ "grad_norm": 0.47112759947776794,
1247
+ "learning_rate": 6.182212581344903e-06,
1248
+ "loss": 0.6315,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.38611713665943603,
1253
+ "grad_norm": 0.5504185557365417,
1254
+ "learning_rate": 6.1605206073752725e-06,
1255
+ "loss": 0.6318,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.3882863340563991,
1260
+ "grad_norm": 0.5567037463188171,
1261
+ "learning_rate": 6.13882863340564e-06,
1262
+ "loss": 0.6104,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.39045553145336226,
1267
+ "grad_norm": 0.4654327630996704,
1268
+ "learning_rate": 6.11713665943601e-06,
1269
+ "loss": 0.619,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.3926247288503254,
1274
+ "grad_norm": 0.47351258993148804,
1275
+ "learning_rate": 6.095444685466378e-06,
1276
+ "loss": 0.6206,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.3947939262472885,
1281
+ "grad_norm": 0.47146716713905334,
1282
+ "learning_rate": 6.073752711496746e-06,
1283
+ "loss": 0.6111,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.3969631236442516,
1288
+ "grad_norm": 0.49498575925827026,
1289
+ "learning_rate": 6.052060737527116e-06,
1290
+ "loss": 0.6179,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.39913232104121477,
1295
+ "grad_norm": 0.5393364429473877,
1296
+ "learning_rate": 6.030368763557483e-06,
1297
+ "loss": 0.6331,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.40130151843817785,
1302
+ "grad_norm": 0.5126506686210632,
1303
+ "learning_rate": 6.008676789587853e-06,
1304
+ "loss": 0.6143,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.403470715835141,
1309
+ "grad_norm": 0.6334784626960754,
1310
+ "learning_rate": 5.986984815618222e-06,
1311
+ "loss": 0.6322,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.40563991323210413,
1316
+ "grad_norm": 0.5737828016281128,
1317
+ "learning_rate": 5.96529284164859e-06,
1318
+ "loss": 0.6206,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.4078091106290672,
1323
+ "grad_norm": 0.5050855875015259,
1324
+ "learning_rate": 5.943600867678959e-06,
1325
+ "loss": 0.6262,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.40997830802603036,
1330
+ "grad_norm": 0.4655757546424866,
1331
+ "learning_rate": 5.921908893709328e-06,
1332
+ "loss": 0.6062,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.4121475054229935,
1337
+ "grad_norm": 0.7657625675201416,
1338
+ "learning_rate": 5.900216919739696e-06,
1339
+ "loss": 0.6278,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.41431670281995664,
1344
+ "grad_norm": 0.4740203619003296,
1345
+ "learning_rate": 5.8785249457700655e-06,
1346
+ "loss": 0.6125,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.4164859002169197,
1351
+ "grad_norm": 0.47104138135910034,
1352
+ "learning_rate": 5.856832971800435e-06,
1353
+ "loss": 0.6181,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.41865509761388287,
1358
+ "grad_norm": 0.6184363961219788,
1359
+ "learning_rate": 5.835140997830803e-06,
1360
+ "loss": 0.6227,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.420824295010846,
1365
+ "grad_norm": 0.6750501990318298,
1366
+ "learning_rate": 5.813449023861172e-06,
1367
+ "loss": 0.6407,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.4229934924078091,
1372
+ "grad_norm": 0.6018952131271362,
1373
+ "learning_rate": 5.791757049891541e-06,
1374
+ "loss": 0.5914,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.42516268980477223,
1379
+ "grad_norm": 0.5619893670082092,
1380
+ "learning_rate": 5.770065075921909e-06,
1381
+ "loss": 0.6222,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.42733188720173537,
1386
+ "grad_norm": 0.4996378719806671,
1387
+ "learning_rate": 5.748373101952278e-06,
1388
+ "loss": 0.6076,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.42950108459869846,
1393
+ "grad_norm": 0.47961318492889404,
1394
+ "learning_rate": 5.7266811279826476e-06,
1395
+ "loss": 0.6102,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.4316702819956616,
1400
+ "grad_norm": 0.6786702871322632,
1401
+ "learning_rate": 5.704989154013015e-06,
1402
+ "loss": 0.6353,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.43383947939262474,
1407
+ "grad_norm": 0.6170738935470581,
1408
+ "learning_rate": 5.683297180043385e-06,
1409
+ "loss": 0.6038,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 0.4360086767895879,
1414
+ "grad_norm": 0.5487593412399292,
1415
+ "learning_rate": 5.6616052060737535e-06,
1416
+ "loss": 0.5929,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 0.43817787418655096,
1421
+ "grad_norm": 0.593626856803894,
1422
+ "learning_rate": 5.639913232104122e-06,
1423
+ "loss": 0.5977,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 0.4403470715835141,
1428
+ "grad_norm": 0.7302510738372803,
1429
+ "learning_rate": 5.618221258134491e-06,
1430
+ "loss": 0.5996,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 0.44251626898047725,
1435
+ "grad_norm": 0.5379276275634766,
1436
+ "learning_rate": 5.59652928416486e-06,
1437
+ "loss": 0.6119,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 0.44468546637744033,
1442
+ "grad_norm": 0.6861233711242676,
1443
+ "learning_rate": 5.574837310195228e-06,
1444
+ "loss": 0.6119,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 0.44685466377440347,
1449
+ "grad_norm": 0.5536229014396667,
1450
+ "learning_rate": 5.5531453362255974e-06,
1451
+ "loss": 0.5926,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 0.4490238611713666,
1456
+ "grad_norm": 0.5160278677940369,
1457
+ "learning_rate": 5.531453362255966e-06,
1458
+ "loss": 0.6013,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 0.4511930585683297,
1463
+ "grad_norm": 0.5844294428825378,
1464
+ "learning_rate": 5.509761388286335e-06,
1465
+ "loss": 0.6044,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 0.45336225596529284,
1470
+ "grad_norm": 0.6622424721717834,
1471
+ "learning_rate": 5.488069414316703e-06,
1472
+ "loss": 0.6005,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 0.455531453362256,
1477
+ "grad_norm": 0.5025097727775574,
1478
+ "learning_rate": 5.466377440347071e-06,
1479
+ "loss": 0.5802,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 0.45770065075921906,
1484
+ "grad_norm": 0.5540115237236023,
1485
+ "learning_rate": 5.444685466377441e-06,
1486
+ "loss": 0.5932,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 0.4598698481561822,
1491
+ "grad_norm": 0.5804458856582642,
1492
+ "learning_rate": 5.42299349240781e-06,
1493
+ "loss": 0.5987,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 0.46203904555314534,
1498
+ "grad_norm": 0.5122389793395996,
1499
+ "learning_rate": 5.401301518438178e-06,
1500
+ "loss": 0.5916,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 0.4642082429501085,
1505
+ "grad_norm": 0.5717864632606506,
1506
+ "learning_rate": 5.379609544468547e-06,
1507
+ "loss": 0.5948,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 0.46637744034707157,
1512
+ "grad_norm": 0.6112474799156189,
1513
+ "learning_rate": 5.357917570498916e-06,
1514
+ "loss": 0.6016,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 0.4685466377440347,
1519
+ "grad_norm": 0.58445805311203,
1520
+ "learning_rate": 5.3362255965292846e-06,
1521
+ "loss": 0.5943,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 0.47071583514099785,
1526
+ "grad_norm": 0.5126850605010986,
1527
+ "learning_rate": 5.314533622559653e-06,
1528
+ "loss": 0.6058,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 0.47288503253796094,
1533
+ "grad_norm": 0.498695969581604,
1534
+ "learning_rate": 5.292841648590023e-06,
1535
+ "loss": 0.5867,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 0.4750542299349241,
1540
+ "grad_norm": 0.5060663223266602,
1541
+ "learning_rate": 5.2711496746203904e-06,
1542
+ "loss": 0.6161,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 0.4772234273318872,
1547
+ "grad_norm": 0.49926185607910156,
1548
+ "learning_rate": 5.24945770065076e-06,
1549
+ "loss": 0.594,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 0.4793926247288503,
1554
+ "grad_norm": 0.7231149673461914,
1555
+ "learning_rate": 5.2277657266811285e-06,
1556
+ "loss": 0.6506,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 0.48156182212581344,
1561
+ "grad_norm": 0.501641571521759,
1562
+ "learning_rate": 5.206073752711497e-06,
1563
+ "loss": 0.6031,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 0.4837310195227766,
1568
+ "grad_norm": 0.5351401567459106,
1569
+ "learning_rate": 5.184381778741866e-06,
1570
+ "loss": 0.5957,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 0.48590021691973967,
1575
+ "grad_norm": 0.4861268997192383,
1576
+ "learning_rate": 5.162689804772235e-06,
1577
+ "loss": 0.6109,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 0.4880694143167028,
1582
+ "grad_norm": 0.4920121133327484,
1583
+ "learning_rate": 5.140997830802603e-06,
1584
+ "loss": 0.5931,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 0.49023861171366595,
1589
+ "grad_norm": 0.5210174322128296,
1590
+ "learning_rate": 5.1193058568329725e-06,
1591
+ "loss": 0.6122,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 0.4924078091106291,
1596
+ "grad_norm": 0.7002459764480591,
1597
+ "learning_rate": 5.097613882863341e-06,
1598
+ "loss": 0.603,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 0.4945770065075922,
1603
+ "grad_norm": 0.4967700242996216,
1604
+ "learning_rate": 5.07592190889371e-06,
1605
+ "loss": 0.5675,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 0.4967462039045553,
1610
+ "grad_norm": 0.49998047947883606,
1611
+ "learning_rate": 5.054229934924078e-06,
1612
+ "loss": 0.5885,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 0.49891540130151846,
1617
+ "grad_norm": 0.46569979190826416,
1618
+ "learning_rate": 5.032537960954448e-06,
1619
+ "loss": 0.588,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 0.5010845986984815,
1624
+ "grad_norm": 0.518514096736908,
1625
+ "learning_rate": 5.010845986984816e-06,
1626
+ "loss": 0.5795,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 0.5032537960954447,
1631
+ "grad_norm": 0.5044274926185608,
1632
+ "learning_rate": 4.989154013015185e-06,
1633
+ "loss": 0.5777,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 0.5054229934924078,
1638
+ "grad_norm": 0.5333325862884521,
1639
+ "learning_rate": 4.967462039045554e-06,
1640
+ "loss": 0.5902,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 0.5075921908893709,
1645
+ "grad_norm": 0.666131854057312,
1646
+ "learning_rate": 4.945770065075922e-06,
1647
+ "loss": 0.5735,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 0.5097613882863341,
1652
+ "grad_norm": 0.5070677995681763,
1653
+ "learning_rate": 4.924078091106291e-06,
1654
+ "loss": 0.5677,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 0.5119305856832972,
1659
+ "grad_norm": 0.5434484481811523,
1660
+ "learning_rate": 4.90238611713666e-06,
1661
+ "loss": 0.5984,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 0.5140997830802603,
1666
+ "grad_norm": 0.5604612231254578,
1667
+ "learning_rate": 4.880694143167028e-06,
1668
+ "loss": 0.5915,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 0.5162689804772235,
1673
+ "grad_norm": 0.5144227147102356,
1674
+ "learning_rate": 4.859002169197397e-06,
1675
+ "loss": 0.5765,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 0.5184381778741866,
1680
+ "grad_norm": 0.4857323169708252,
1681
+ "learning_rate": 4.837310195227766e-06,
1682
+ "loss": 0.5722,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 0.5206073752711496,
1687
+ "grad_norm": 0.4755032956600189,
1688
+ "learning_rate": 4.815618221258135e-06,
1689
+ "loss": 0.5749,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 0.5227765726681128,
1694
+ "grad_norm": 0.587493896484375,
1695
+ "learning_rate": 4.793926247288504e-06,
1696
+ "loss": 0.5981,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 0.5249457700650759,
1701
+ "grad_norm": 0.5679769515991211,
1702
+ "learning_rate": 4.772234273318872e-06,
1703
+ "loss": 0.5884,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 0.527114967462039,
1708
+ "grad_norm": 0.498691588640213,
1709
+ "learning_rate": 4.750542299349241e-06,
1710
+ "loss": 0.5809,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 0.5292841648590022,
1715
+ "grad_norm": 0.675168514251709,
1716
+ "learning_rate": 4.7288503253796095e-06,
1717
+ "loss": 0.5874,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 0.5314533622559653,
1722
+ "grad_norm": 0.5249897241592407,
1723
+ "learning_rate": 4.707158351409979e-06,
1724
+ "loss": 0.5878,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 0.5336225596529284,
1729
+ "grad_norm": 0.48491179943084717,
1730
+ "learning_rate": 4.685466377440348e-06,
1731
+ "loss": 0.5743,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 0.5357917570498916,
1736
+ "grad_norm": 0.5798050761222839,
1737
+ "learning_rate": 4.663774403470716e-06,
1738
+ "loss": 0.5682,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 0.5379609544468547,
1743
+ "grad_norm": 0.5193563103675842,
1744
+ "learning_rate": 4.642082429501085e-06,
1745
+ "loss": 0.5743,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 0.5401301518438177,
1750
+ "grad_norm": 0.5050124526023865,
1751
+ "learning_rate": 4.6203904555314535e-06,
1752
+ "loss": 0.5823,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 0.5422993492407809,
1757
+ "grad_norm": 0.5084236264228821,
1758
+ "learning_rate": 4.598698481561822e-06,
1759
+ "loss": 0.5865,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 0.544468546637744,
1764
+ "grad_norm": 0.5621696710586548,
1765
+ "learning_rate": 4.577006507592191e-06,
1766
+ "loss": 0.6023,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 0.5466377440347071,
1771
+ "grad_norm": 0.4698992669582367,
1772
+ "learning_rate": 4.55531453362256e-06,
1773
+ "loss": 0.5835,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 0.5488069414316703,
1778
+ "grad_norm": 0.5616809129714966,
1779
+ "learning_rate": 4.533622559652929e-06,
1780
+ "loss": 0.559,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 0.5509761388286334,
1785
+ "grad_norm": 0.5029488801956177,
1786
+ "learning_rate": 4.5119305856832975e-06,
1787
+ "loss": 0.5693,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 0.5531453362255966,
1792
+ "grad_norm": 0.47164446115493774,
1793
+ "learning_rate": 4.490238611713666e-06,
1794
+ "loss": 0.564,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 0.5553145336225597,
1799
+ "grad_norm": 0.49710217118263245,
1800
+ "learning_rate": 4.468546637744035e-06,
1801
+ "loss": 0.5826,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 0.5574837310195228,
1806
+ "grad_norm": 0.48587456345558167,
1807
+ "learning_rate": 4.446854663774403e-06,
1808
+ "loss": 0.5816,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 0.559652928416486,
1813
+ "grad_norm": 0.5498842597007751,
1814
+ "learning_rate": 4.425162689804773e-06,
1815
+ "loss": 0.5736,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 0.561822125813449,
1820
+ "grad_norm": 0.4747694432735443,
1821
+ "learning_rate": 4.4034707158351415e-06,
1822
+ "loss": 0.574,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 0.5639913232104121,
1827
+ "grad_norm": 0.9344505667686462,
1828
+ "learning_rate": 4.38177874186551e-06,
1829
+ "loss": 0.5957,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 0.5661605206073753,
1834
+ "grad_norm": 0.5000290274620056,
1835
+ "learning_rate": 4.360086767895879e-06,
1836
+ "loss": 0.5774,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 0.5683297180043384,
1841
+ "grad_norm": 0.5080724358558655,
1842
+ "learning_rate": 4.338394793926247e-06,
1843
+ "loss": 0.5688,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 0.5704989154013015,
1848
+ "grad_norm": 0.5054919123649597,
1849
+ "learning_rate": 4.316702819956616e-06,
1850
+ "loss": 0.5834,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 0.5726681127982647,
1855
+ "grad_norm": 0.5046693682670593,
1856
+ "learning_rate": 4.295010845986985e-06,
1857
+ "loss": 0.5687,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 0.5748373101952278,
1862
+ "grad_norm": 0.5170451998710632,
1863
+ "learning_rate": 4.273318872017354e-06,
1864
+ "loss": 0.5497,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 0.5770065075921909,
1869
+ "grad_norm": 0.5297961235046387,
1870
+ "learning_rate": 4.251626898047723e-06,
1871
+ "loss": 0.6013,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 0.579175704989154,
1876
+ "grad_norm": 0.4973108172416687,
1877
+ "learning_rate": 4.229934924078091e-06,
1878
+ "loss": 0.5544,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 0.5813449023861171,
1883
+ "grad_norm": 0.6242494583129883,
1884
+ "learning_rate": 4.208242950108461e-06,
1885
+ "loss": 0.5917,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 0.5835140997830802,
1890
+ "grad_norm": 0.5227379202842712,
1891
+ "learning_rate": 4.186550976138829e-06,
1892
+ "loss": 0.5709,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 0.5856832971800434,
1897
+ "grad_norm": 0.527258574962616,
1898
+ "learning_rate": 4.164859002169197e-06,
1899
+ "loss": 0.5746,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 0.5878524945770065,
1904
+ "grad_norm": 0.5078297257423401,
1905
+ "learning_rate": 4.143167028199567e-06,
1906
+ "loss": 0.5569,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 0.5900216919739696,
1911
+ "grad_norm": 0.7073683738708496,
1912
+ "learning_rate": 4.121475054229935e-06,
1913
+ "loss": 0.5806,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 0.5921908893709328,
1918
+ "grad_norm": 0.6542203426361084,
1919
+ "learning_rate": 4.099783080260304e-06,
1920
+ "loss": 0.5963,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 0.5943600867678959,
1925
+ "grad_norm": 0.4986463189125061,
1926
+ "learning_rate": 4.078091106290673e-06,
1927
+ "loss": 0.5786,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 0.596529284164859,
1932
+ "grad_norm": 0.4862741231918335,
1933
+ "learning_rate": 4.056399132321042e-06,
1934
+ "loss": 0.5508,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 0.5986984815618221,
1939
+ "grad_norm": 0.5407434105873108,
1940
+ "learning_rate": 4.03470715835141e-06,
1941
+ "loss": 0.5586,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 0.6008676789587852,
1946
+ "grad_norm": 0.513647198677063,
1947
+ "learning_rate": 4.0130151843817785e-06,
1948
+ "loss": 0.5653,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 0.6030368763557483,
1953
+ "grad_norm": 0.4941583573818207,
1954
+ "learning_rate": 3.991323210412148e-06,
1955
+ "loss": 0.5724,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 0.6052060737527115,
1960
+ "grad_norm": 0.5022282600402832,
1961
+ "learning_rate": 3.9696312364425166e-06,
1962
+ "loss": 0.5597,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 0.6073752711496746,
1967
+ "grad_norm": 0.5344776511192322,
1968
+ "learning_rate": 3.947939262472885e-06,
1969
+ "loss": 0.576,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 0.6095444685466378,
1974
+ "grad_norm": 0.8711801171302795,
1975
+ "learning_rate": 3.926247288503255e-06,
1976
+ "loss": 0.5653,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 0.6117136659436009,
1981
+ "grad_norm": 0.6406508684158325,
1982
+ "learning_rate": 3.904555314533623e-06,
1983
+ "loss": 0.5612,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 0.613882863340564,
1988
+ "grad_norm": 0.5289674997329712,
1989
+ "learning_rate": 3.882863340563991e-06,
1990
+ "loss": 0.5553,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 0.6160520607375272,
1995
+ "grad_norm": 0.49930331110954285,
1996
+ "learning_rate": 3.8611713665943606e-06,
1997
+ "loss": 0.5661,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 0.6182212581344902,
2002
+ "grad_norm": 0.49601882696151733,
2003
+ "learning_rate": 3.839479392624729e-06,
2004
+ "loss": 0.5613,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 0.6203904555314533,
2009
+ "grad_norm": 0.5551280975341797,
2010
+ "learning_rate": 3.817787418655098e-06,
2011
+ "loss": 0.5692,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 0.6225596529284165,
2016
+ "grad_norm": 0.5268258452415466,
2017
+ "learning_rate": 3.7960954446854664e-06,
2018
+ "loss": 0.5653,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 0.6247288503253796,
2023
+ "grad_norm": 0.5202245712280273,
2024
+ "learning_rate": 3.7744034707158355e-06,
2025
+ "loss": 0.5623,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 0.6268980477223427,
2030
+ "grad_norm": 0.7391339540481567,
2031
+ "learning_rate": 3.752711496746204e-06,
2032
+ "loss": 0.5485,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 0.6290672451193059,
2037
+ "grad_norm": 0.7942306995391846,
2038
+ "learning_rate": 3.7310195227765728e-06,
2039
+ "loss": 0.5693,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 0.631236442516269,
2044
+ "grad_norm": 0.5168442130088806,
2045
+ "learning_rate": 3.709327548806942e-06,
2046
+ "loss": 0.563,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 0.6334056399132321,
2051
+ "grad_norm": 0.5481964349746704,
2052
+ "learning_rate": 3.6876355748373104e-06,
2053
+ "loss": 0.5694,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 0.6355748373101953,
2058
+ "grad_norm": 0.7699911594390869,
2059
+ "learning_rate": 3.665943600867679e-06,
2060
+ "loss": 0.602,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 0.6377440347071583,
2065
+ "grad_norm": 0.5012184381484985,
2066
+ "learning_rate": 3.644251626898048e-06,
2067
+ "loss": 0.5566,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 0.6399132321041214,
2072
+ "grad_norm": 0.5150504112243652,
2073
+ "learning_rate": 3.6225596529284167e-06,
2074
+ "loss": 0.5575,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 0.6420824295010846,
2079
+ "grad_norm": 0.4825197160243988,
2080
+ "learning_rate": 3.6008676789587854e-06,
2081
+ "loss": 0.578,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 0.6442516268980477,
2086
+ "grad_norm": 0.5708716511726379,
2087
+ "learning_rate": 3.5791757049891544e-06,
2088
+ "loss": 0.5748,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 0.6464208242950108,
2093
+ "grad_norm": 0.49083611369132996,
2094
+ "learning_rate": 3.557483731019523e-06,
2095
+ "loss": 0.5656,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 0.648590021691974,
2100
+ "grad_norm": 0.5466710329055786,
2101
+ "learning_rate": 3.5357917570498917e-06,
2102
+ "loss": 0.5808,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 0.6507592190889371,
2107
+ "grad_norm": 0.6208142042160034,
2108
+ "learning_rate": 3.5140997830802603e-06,
2109
+ "loss": 0.5665,
2110
+ "step": 300
2111
+ },
2112
+ {
2113
+ "epoch": 0.6529284164859002,
2114
+ "grad_norm": 0.4832100570201874,
2115
+ "learning_rate": 3.4924078091106293e-06,
2116
+ "loss": 0.572,
2117
+ "step": 301
2118
+ },
2119
+ {
2120
+ "epoch": 0.6550976138828634,
2121
+ "grad_norm": 0.5140089392662048,
2122
+ "learning_rate": 3.470715835140998e-06,
2123
+ "loss": 0.5565,
2124
+ "step": 302
2125
+ },
2126
+ {
2127
+ "epoch": 0.6572668112798264,
2128
+ "grad_norm": 0.5033791661262512,
2129
+ "learning_rate": 3.4490238611713666e-06,
2130
+ "loss": 0.5522,
2131
+ "step": 303
2132
+ },
2133
+ {
2134
+ "epoch": 0.6594360086767896,
2135
+ "grad_norm": 0.4576267600059509,
2136
+ "learning_rate": 3.4273318872017357e-06,
2137
+ "loss": 0.5511,
2138
+ "step": 304
2139
+ },
2140
+ {
2141
+ "epoch": 0.6616052060737527,
2142
+ "grad_norm": 0.5113222002983093,
2143
+ "learning_rate": 3.4056399132321043e-06,
2144
+ "loss": 0.5736,
2145
+ "step": 305
2146
+ },
2147
+ {
2148
+ "epoch": 0.6637744034707158,
2149
+ "grad_norm": 0.528408408164978,
2150
+ "learning_rate": 3.383947939262473e-06,
2151
+ "loss": 0.5679,
2152
+ "step": 306
2153
+ },
2154
+ {
2155
+ "epoch": 0.665943600867679,
2156
+ "grad_norm": 0.4761233925819397,
2157
+ "learning_rate": 3.362255965292842e-06,
2158
+ "loss": 0.535,
2159
+ "step": 307
2160
+ },
2161
+ {
2162
+ "epoch": 0.6681127982646421,
2163
+ "grad_norm": 0.4666268527507782,
2164
+ "learning_rate": 3.3405639913232106e-06,
2165
+ "loss": 0.5731,
2166
+ "step": 308
2167
+ },
2168
+ {
2169
+ "epoch": 0.6702819956616052,
2170
+ "grad_norm": 0.5518378615379333,
2171
+ "learning_rate": 3.3188720173535792e-06,
2172
+ "loss": 0.5679,
2173
+ "step": 309
2174
+ },
2175
+ {
2176
+ "epoch": 0.6724511930585684,
2177
+ "grad_norm": 0.4833706021308899,
2178
+ "learning_rate": 3.2971800433839487e-06,
2179
+ "loss": 0.5615,
2180
+ "step": 310
2181
+ },
2182
+ {
2183
+ "epoch": 0.6746203904555315,
2184
+ "grad_norm": 0.5420160889625549,
2185
+ "learning_rate": 3.275488069414317e-06,
2186
+ "loss": 0.5529,
2187
+ "step": 311
2188
+ },
2189
+ {
2190
+ "epoch": 0.6767895878524945,
2191
+ "grad_norm": 0.6026318669319153,
2192
+ "learning_rate": 3.2537960954446855e-06,
2193
+ "loss": 0.5478,
2194
+ "step": 312
2195
+ },
2196
+ {
2197
+ "epoch": 0.6789587852494577,
2198
+ "grad_norm": 0.5486297607421875,
2199
+ "learning_rate": 3.232104121475054e-06,
2200
+ "loss": 0.5619,
2201
+ "step": 313
2202
+ },
2203
+ {
2204
+ "epoch": 0.6811279826464208,
2205
+ "grad_norm": 0.49950018525123596,
2206
+ "learning_rate": 3.210412147505423e-06,
2207
+ "loss": 0.5539,
2208
+ "step": 314
2209
+ },
2210
+ {
2211
+ "epoch": 0.6832971800433839,
2212
+ "grad_norm": 0.4778575301170349,
2213
+ "learning_rate": 3.188720173535792e-06,
2214
+ "loss": 0.546,
2215
+ "step": 315
2216
+ },
2217
+ {
2218
+ "epoch": 0.6854663774403471,
2219
+ "grad_norm": 0.4857572317123413,
2220
+ "learning_rate": 3.1670281995661605e-06,
2221
+ "loss": 0.5502,
2222
+ "step": 316
2223
+ },
2224
+ {
2225
+ "epoch": 0.6876355748373102,
2226
+ "grad_norm": 0.5437939167022705,
2227
+ "learning_rate": 3.14533622559653e-06,
2228
+ "loss": 0.5448,
2229
+ "step": 317
2230
+ },
2231
+ {
2232
+ "epoch": 0.6898047722342733,
2233
+ "grad_norm": 0.6015022397041321,
2234
+ "learning_rate": 3.123644251626898e-06,
2235
+ "loss": 0.5562,
2236
+ "step": 318
2237
+ },
2238
+ {
2239
+ "epoch": 0.6919739696312365,
2240
+ "grad_norm": 0.5425083041191101,
2241
+ "learning_rate": 3.1019522776572668e-06,
2242
+ "loss": 0.5496,
2243
+ "step": 319
2244
+ },
2245
+ {
2246
+ "epoch": 0.6941431670281996,
2247
+ "grad_norm": 0.4970496594905853,
2248
+ "learning_rate": 3.0802603036876362e-06,
2249
+ "loss": 0.5577,
2250
+ "step": 320
2251
+ },
2252
+ {
2253
+ "epoch": 0.6963123644251626,
2254
+ "grad_norm": 0.5393553972244263,
2255
+ "learning_rate": 3.058568329718005e-06,
2256
+ "loss": 0.5665,
2257
+ "step": 321
2258
+ },
2259
+ {
2260
+ "epoch": 0.6984815618221258,
2261
+ "grad_norm": 0.5788228511810303,
2262
+ "learning_rate": 3.036876355748373e-06,
2263
+ "loss": 0.5525,
2264
+ "step": 322
2265
+ },
2266
+ {
2267
+ "epoch": 0.7006507592190889,
2268
+ "grad_norm": 0.582302987575531,
2269
+ "learning_rate": 3.0151843817787417e-06,
2270
+ "loss": 0.5337,
2271
+ "step": 323
2272
+ },
2273
+ {
2274
+ "epoch": 0.702819956616052,
2275
+ "grad_norm": 0.5455271601676941,
2276
+ "learning_rate": 2.993492407809111e-06,
2277
+ "loss": 0.567,
2278
+ "step": 324
2279
+ },
2280
+ {
2281
+ "epoch": 0.7049891540130152,
2282
+ "grad_norm": 0.5106908679008484,
2283
+ "learning_rate": 2.9718004338394794e-06,
2284
+ "loss": 0.5572,
2285
+ "step": 325
2286
+ },
2287
+ {
2288
+ "epoch": 0.7071583514099783,
2289
+ "grad_norm": 0.5830085277557373,
2290
+ "learning_rate": 2.950108459869848e-06,
2291
+ "loss": 0.5799,
2292
+ "step": 326
2293
+ },
2294
+ {
2295
+ "epoch": 0.7093275488069414,
2296
+ "grad_norm": 0.543565571308136,
2297
+ "learning_rate": 2.9284164859002175e-06,
2298
+ "loss": 0.5486,
2299
+ "step": 327
2300
+ },
2301
+ {
2302
+ "epoch": 0.7114967462039046,
2303
+ "grad_norm": 0.6363551616668701,
2304
+ "learning_rate": 2.906724511930586e-06,
2305
+ "loss": 0.5589,
2306
+ "step": 328
2307
+ },
2308
+ {
2309
+ "epoch": 0.7136659436008677,
2310
+ "grad_norm": 0.5492388606071472,
2311
+ "learning_rate": 2.8850325379609543e-06,
2312
+ "loss": 0.5439,
2313
+ "step": 329
2314
+ },
2315
+ {
2316
+ "epoch": 0.7158351409978309,
2317
+ "grad_norm": 0.48855558037757874,
2318
+ "learning_rate": 2.8633405639913238e-06,
2319
+ "loss": 0.5342,
2320
+ "step": 330
2321
+ },
2322
+ {
2323
+ "epoch": 0.7180043383947939,
2324
+ "grad_norm": 0.5338431000709534,
2325
+ "learning_rate": 2.8416485900216924e-06,
2326
+ "loss": 0.5313,
2327
+ "step": 331
2328
+ },
2329
+ {
2330
+ "epoch": 0.720173535791757,
2331
+ "grad_norm": 0.5648824572563171,
2332
+ "learning_rate": 2.819956616052061e-06,
2333
+ "loss": 0.5521,
2334
+ "step": 332
2335
+ },
2336
+ {
2337
+ "epoch": 0.7223427331887202,
2338
+ "grad_norm": 0.5221781730651855,
2339
+ "learning_rate": 2.79826464208243e-06,
2340
+ "loss": 0.5445,
2341
+ "step": 333
2342
+ },
2343
+ {
2344
+ "epoch": 0.7245119305856833,
2345
+ "grad_norm": 0.4889802634716034,
2346
+ "learning_rate": 2.7765726681127987e-06,
2347
+ "loss": 0.5588,
2348
+ "step": 334
2349
+ },
2350
+ {
2351
+ "epoch": 0.7266811279826464,
2352
+ "grad_norm": 0.4981290400028229,
2353
+ "learning_rate": 2.7548806941431673e-06,
2354
+ "loss": 0.5709,
2355
+ "step": 335
2356
+ },
2357
+ {
2358
+ "epoch": 0.7288503253796096,
2359
+ "grad_norm": 0.5129498243331909,
2360
+ "learning_rate": 2.7331887201735356e-06,
2361
+ "loss": 0.5592,
2362
+ "step": 336
2363
+ },
2364
+ {
2365
+ "epoch": 0.7310195227765727,
2366
+ "grad_norm": 0.5097779631614685,
2367
+ "learning_rate": 2.711496746203905e-06,
2368
+ "loss": 0.5608,
2369
+ "step": 337
2370
+ },
2371
+ {
2372
+ "epoch": 0.7331887201735358,
2373
+ "grad_norm": 0.49256396293640137,
2374
+ "learning_rate": 2.6898047722342737e-06,
2375
+ "loss": 0.547,
2376
+ "step": 338
2377
+ },
2378
+ {
2379
+ "epoch": 0.735357917570499,
2380
+ "grad_norm": 0.49484625458717346,
2381
+ "learning_rate": 2.6681127982646423e-06,
2382
+ "loss": 0.5605,
2383
+ "step": 339
2384
+ },
2385
+ {
2386
+ "epoch": 0.737527114967462,
2387
+ "grad_norm": 0.4865284264087677,
2388
+ "learning_rate": 2.6464208242950113e-06,
2389
+ "loss": 0.552,
2390
+ "step": 340
2391
+ },
2392
+ {
2393
+ "epoch": 0.7396963123644251,
2394
+ "grad_norm": 0.5712233185768127,
2395
+ "learning_rate": 2.62472885032538e-06,
2396
+ "loss": 0.5229,
2397
+ "step": 341
2398
+ },
2399
+ {
2400
+ "epoch": 0.7418655097613883,
2401
+ "grad_norm": 0.4782271981239319,
2402
+ "learning_rate": 2.6030368763557486e-06,
2403
+ "loss": 0.5409,
2404
+ "step": 342
2405
+ },
2406
+ {
2407
+ "epoch": 0.7440347071583514,
2408
+ "grad_norm": 0.48679468035697937,
2409
+ "learning_rate": 2.5813449023861176e-06,
2410
+ "loss": 0.5502,
2411
+ "step": 343
2412
+ },
2413
+ {
2414
+ "epoch": 0.7462039045553145,
2415
+ "grad_norm": 0.49135392904281616,
2416
+ "learning_rate": 2.5596529284164863e-06,
2417
+ "loss": 0.552,
2418
+ "step": 344
2419
+ },
2420
+ {
2421
+ "epoch": 0.7483731019522777,
2422
+ "grad_norm": 0.4905093014240265,
2423
+ "learning_rate": 2.537960954446855e-06,
2424
+ "loss": 0.5388,
2425
+ "step": 345
2426
+ },
2427
+ {
2428
+ "epoch": 0.7505422993492408,
2429
+ "grad_norm": 0.4936610758304596,
2430
+ "learning_rate": 2.516268980477224e-06,
2431
+ "loss": 0.5408,
2432
+ "step": 346
2433
+ },
2434
+ {
2435
+ "epoch": 0.7527114967462039,
2436
+ "grad_norm": 0.4581211805343628,
2437
+ "learning_rate": 2.4945770065075926e-06,
2438
+ "loss": 0.5438,
2439
+ "step": 347
2440
+ },
2441
+ {
2442
+ "epoch": 0.754880694143167,
2443
+ "grad_norm": 0.9855303764343262,
2444
+ "learning_rate": 2.472885032537961e-06,
2445
+ "loss": 0.5422,
2446
+ "step": 348
2447
+ },
2448
+ {
2449
+ "epoch": 0.7570498915401301,
2450
+ "grad_norm": 0.4950404465198517,
2451
+ "learning_rate": 2.45119305856833e-06,
2452
+ "loss": 0.5488,
2453
+ "step": 349
2454
+ },
2455
+ {
2456
+ "epoch": 0.7592190889370932,
2457
+ "grad_norm": 0.5436333417892456,
2458
+ "learning_rate": 2.4295010845986985e-06,
2459
+ "loss": 0.529,
2460
+ "step": 350
2461
+ },
2462
+ {
2463
+ "epoch": 0.7613882863340564,
2464
+ "grad_norm": 0.6923372149467468,
2465
+ "learning_rate": 2.4078091106290675e-06,
2466
+ "loss": 0.566,
2467
+ "step": 351
2468
+ },
2469
+ {
2470
+ "epoch": 0.7635574837310195,
2471
+ "grad_norm": 0.4927901327610016,
2472
+ "learning_rate": 2.386117136659436e-06,
2473
+ "loss": 0.5445,
2474
+ "step": 352
2475
+ },
2476
+ {
2477
+ "epoch": 0.7657266811279827,
2478
+ "grad_norm": 0.6745809316635132,
2479
+ "learning_rate": 2.3644251626898048e-06,
2480
+ "loss": 0.5644,
2481
+ "step": 353
2482
+ },
2483
+ {
2484
+ "epoch": 0.7678958785249458,
2485
+ "grad_norm": 0.5441464185714722,
2486
+ "learning_rate": 2.342733188720174e-06,
2487
+ "loss": 0.5481,
2488
+ "step": 354
2489
+ },
2490
+ {
2491
+ "epoch": 0.7700650759219089,
2492
+ "grad_norm": 0.5015382170677185,
2493
+ "learning_rate": 2.3210412147505424e-06,
2494
+ "loss": 0.5632,
2495
+ "step": 355
2496
+ },
2497
+ {
2498
+ "epoch": 0.7722342733188721,
2499
+ "grad_norm": 0.48177745938301086,
2500
+ "learning_rate": 2.299349240780911e-06,
2501
+ "loss": 0.5435,
2502
+ "step": 356
2503
+ },
2504
+ {
2505
+ "epoch": 0.7744034707158352,
2506
+ "grad_norm": 0.6477603316307068,
2507
+ "learning_rate": 2.27765726681128e-06,
2508
+ "loss": 0.5488,
2509
+ "step": 357
2510
+ },
2511
+ {
2512
+ "epoch": 0.7765726681127982,
2513
+ "grad_norm": 0.49363404512405396,
2514
+ "learning_rate": 2.2559652928416487e-06,
2515
+ "loss": 0.5359,
2516
+ "step": 358
2517
+ },
2518
+ {
2519
+ "epoch": 0.7787418655097614,
2520
+ "grad_norm": 0.49064791202545166,
2521
+ "learning_rate": 2.2342733188720174e-06,
2522
+ "loss": 0.5708,
2523
+ "step": 359
2524
+ },
2525
+ {
2526
+ "epoch": 0.7809110629067245,
2527
+ "grad_norm": 0.5506601929664612,
2528
+ "learning_rate": 2.2125813449023864e-06,
2529
+ "loss": 0.5638,
2530
+ "step": 360
2531
+ },
2532
+ {
2533
+ "epoch": 0.7830802603036876,
2534
+ "grad_norm": 0.49600258469581604,
2535
+ "learning_rate": 2.190889370932755e-06,
2536
+ "loss": 0.5578,
2537
+ "step": 361
2538
+ },
2539
+ {
2540
+ "epoch": 0.7852494577006508,
2541
+ "grad_norm": 0.6045665740966797,
2542
+ "learning_rate": 2.1691973969631237e-06,
2543
+ "loss": 0.5387,
2544
+ "step": 362
2545
+ },
2546
+ {
2547
+ "epoch": 0.7874186550976139,
2548
+ "grad_norm": 0.4785294234752655,
2549
+ "learning_rate": 2.1475054229934923e-06,
2550
+ "loss": 0.5461,
2551
+ "step": 363
2552
+ },
2553
+ {
2554
+ "epoch": 0.789587852494577,
2555
+ "grad_norm": 0.505807101726532,
2556
+ "learning_rate": 2.1258134490238614e-06,
2557
+ "loss": 0.5421,
2558
+ "step": 364
2559
+ },
2560
+ {
2561
+ "epoch": 0.7917570498915402,
2562
+ "grad_norm": 0.5481471419334412,
2563
+ "learning_rate": 2.1041214750542304e-06,
2564
+ "loss": 0.5359,
2565
+ "step": 365
2566
+ },
2567
+ {
2568
+ "epoch": 0.7939262472885033,
2569
+ "grad_norm": 0.5260441899299622,
2570
+ "learning_rate": 2.0824295010845986e-06,
2571
+ "loss": 0.5801,
2572
+ "step": 366
2573
+ },
2574
+ {
2575
+ "epoch": 0.7960954446854663,
2576
+ "grad_norm": 0.5684512257575989,
2577
+ "learning_rate": 2.0607375271149677e-06,
2578
+ "loss": 0.5317,
2579
+ "step": 367
2580
+ },
2581
+ {
2582
+ "epoch": 0.7982646420824295,
2583
+ "grad_norm": 0.5323051810264587,
2584
+ "learning_rate": 2.0390455531453363e-06,
2585
+ "loss": 0.5669,
2586
+ "step": 368
2587
+ },
2588
+ {
2589
+ "epoch": 0.8004338394793926,
2590
+ "grad_norm": 0.6611142158508301,
2591
+ "learning_rate": 2.017353579175705e-06,
2592
+ "loss": 0.5582,
2593
+ "step": 369
2594
+ },
2595
+ {
2596
+ "epoch": 0.8026030368763557,
2597
+ "grad_norm": 0.47256988286972046,
2598
+ "learning_rate": 1.995661605206074e-06,
2599
+ "loss": 0.5276,
2600
+ "step": 370
2601
+ },
2602
+ {
2603
+ "epoch": 0.8047722342733189,
2604
+ "grad_norm": 0.6162890195846558,
2605
+ "learning_rate": 1.9739696312364426e-06,
2606
+ "loss": 0.5517,
2607
+ "step": 371
2608
+ },
2609
+ {
2610
+ "epoch": 0.806941431670282,
2611
+ "grad_norm": 0.5171302556991577,
2612
+ "learning_rate": 1.9522776572668117e-06,
2613
+ "loss": 0.5467,
2614
+ "step": 372
2615
+ },
2616
+ {
2617
+ "epoch": 0.8091106290672451,
2618
+ "grad_norm": 0.5168177485466003,
2619
+ "learning_rate": 1.9305856832971803e-06,
2620
+ "loss": 0.537,
2621
+ "step": 373
2622
+ },
2623
+ {
2624
+ "epoch": 0.8112798264642083,
2625
+ "grad_norm": 0.49433693289756775,
2626
+ "learning_rate": 1.908893709327549e-06,
2627
+ "loss": 0.5441,
2628
+ "step": 374
2629
+ },
2630
+ {
2631
+ "epoch": 0.8134490238611713,
2632
+ "grad_norm": 0.6254696249961853,
2633
+ "learning_rate": 1.8872017353579177e-06,
2634
+ "loss": 0.568,
2635
+ "step": 375
2636
+ },
2637
+ {
2638
+ "epoch": 0.8156182212581344,
2639
+ "grad_norm": 0.4846223294734955,
2640
+ "learning_rate": 1.8655097613882864e-06,
2641
+ "loss": 0.5314,
2642
+ "step": 376
2643
+ },
2644
+ {
2645
+ "epoch": 0.8177874186550976,
2646
+ "grad_norm": 0.5695744156837463,
2647
+ "learning_rate": 1.8438177874186552e-06,
2648
+ "loss": 0.5556,
2649
+ "step": 377
2650
+ },
2651
+ {
2652
+ "epoch": 0.8199566160520607,
2653
+ "grad_norm": 0.5135949850082397,
2654
+ "learning_rate": 1.822125813449024e-06,
2655
+ "loss": 0.5513,
2656
+ "step": 378
2657
+ },
2658
+ {
2659
+ "epoch": 0.8221258134490239,
2660
+ "grad_norm": 0.5039001107215881,
2661
+ "learning_rate": 1.8004338394793927e-06,
2662
+ "loss": 0.5395,
2663
+ "step": 379
2664
+ },
2665
+ {
2666
+ "epoch": 0.824295010845987,
2667
+ "grad_norm": 0.5153144598007202,
2668
+ "learning_rate": 1.7787418655097615e-06,
2669
+ "loss": 0.5499,
2670
+ "step": 380
2671
+ },
2672
+ {
2673
+ "epoch": 0.8264642082429501,
2674
+ "grad_norm": 0.6042090058326721,
2675
+ "learning_rate": 1.7570498915401301e-06,
2676
+ "loss": 0.561,
2677
+ "step": 381
2678
+ },
2679
+ {
2680
+ "epoch": 0.8286334056399133,
2681
+ "grad_norm": 0.5201963186264038,
2682
+ "learning_rate": 1.735357917570499e-06,
2683
+ "loss": 0.5514,
2684
+ "step": 382
2685
+ },
2686
+ {
2687
+ "epoch": 0.8308026030368764,
2688
+ "grad_norm": 0.7226424217224121,
2689
+ "learning_rate": 1.7136659436008678e-06,
2690
+ "loss": 0.5481,
2691
+ "step": 383
2692
+ },
2693
+ {
2694
+ "epoch": 0.8329718004338394,
2695
+ "grad_norm": 0.49008411169052124,
2696
+ "learning_rate": 1.6919739696312365e-06,
2697
+ "loss": 0.5577,
2698
+ "step": 384
2699
+ },
2700
+ {
2701
+ "epoch": 0.8351409978308026,
2702
+ "grad_norm": 0.5384876728057861,
2703
+ "learning_rate": 1.6702819956616053e-06,
2704
+ "loss": 0.5744,
2705
+ "step": 385
2706
+ },
2707
+ {
2708
+ "epoch": 0.8373101952277657,
2709
+ "grad_norm": 0.6859086751937866,
2710
+ "learning_rate": 1.6485900216919743e-06,
2711
+ "loss": 0.5495,
2712
+ "step": 386
2713
+ },
2714
+ {
2715
+ "epoch": 0.8394793926247288,
2716
+ "grad_norm": 0.5295915007591248,
2717
+ "learning_rate": 1.6268980477223428e-06,
2718
+ "loss": 0.5555,
2719
+ "step": 387
2720
+ },
2721
+ {
2722
+ "epoch": 0.841648590021692,
2723
+ "grad_norm": 0.49512577056884766,
2724
+ "learning_rate": 1.6052060737527116e-06,
2725
+ "loss": 0.5336,
2726
+ "step": 388
2727
+ },
2728
+ {
2729
+ "epoch": 0.8438177874186551,
2730
+ "grad_norm": 0.5301417112350464,
2731
+ "learning_rate": 1.5835140997830802e-06,
2732
+ "loss": 0.5366,
2733
+ "step": 389
2734
+ },
2735
+ {
2736
+ "epoch": 0.8459869848156182,
2737
+ "grad_norm": 0.5286414623260498,
2738
+ "learning_rate": 1.561822125813449e-06,
2739
+ "loss": 0.5301,
2740
+ "step": 390
2741
+ },
2742
+ {
2743
+ "epoch": 0.8481561822125814,
2744
+ "grad_norm": 0.4798189699649811,
2745
+ "learning_rate": 1.5401301518438181e-06,
2746
+ "loss": 0.5529,
2747
+ "step": 391
2748
+ },
2749
+ {
2750
+ "epoch": 0.8503253796095445,
2751
+ "grad_norm": 0.5175362825393677,
2752
+ "learning_rate": 1.5184381778741865e-06,
2753
+ "loss": 0.5418,
2754
+ "step": 392
2755
+ },
2756
+ {
2757
+ "epoch": 0.8524945770065075,
2758
+ "grad_norm": 0.5054755210876465,
2759
+ "learning_rate": 1.4967462039045556e-06,
2760
+ "loss": 0.5471,
2761
+ "step": 393
2762
+ },
2763
+ {
2764
+ "epoch": 0.8546637744034707,
2765
+ "grad_norm": 0.48426923155784607,
2766
+ "learning_rate": 1.475054229934924e-06,
2767
+ "loss": 0.543,
2768
+ "step": 394
2769
+ },
2770
+ {
2771
+ "epoch": 0.8568329718004338,
2772
+ "grad_norm": 0.49632081389427185,
2773
+ "learning_rate": 1.453362255965293e-06,
2774
+ "loss": 0.5336,
2775
+ "step": 395
2776
+ },
2777
+ {
2778
+ "epoch": 0.8590021691973969,
2779
+ "grad_norm": 0.5040681958198547,
2780
+ "learning_rate": 1.4316702819956619e-06,
2781
+ "loss": 0.5334,
2782
+ "step": 396
2783
+ },
2784
+ {
2785
+ "epoch": 0.8611713665943601,
2786
+ "grad_norm": 0.5700734853744507,
2787
+ "learning_rate": 1.4099783080260305e-06,
2788
+ "loss": 0.5365,
2789
+ "step": 397
2790
+ },
2791
+ {
2792
+ "epoch": 0.8633405639913232,
2793
+ "grad_norm": 0.5232958197593689,
2794
+ "learning_rate": 1.3882863340563994e-06,
2795
+ "loss": 0.5372,
2796
+ "step": 398
2797
+ },
2798
+ {
2799
+ "epoch": 0.8655097613882863,
2800
+ "grad_norm": 0.5089038610458374,
2801
+ "learning_rate": 1.3665943600867678e-06,
2802
+ "loss": 0.5431,
2803
+ "step": 399
2804
+ },
2805
+ {
2806
+ "epoch": 0.8676789587852495,
2807
+ "grad_norm": 0.5343663692474365,
2808
+ "learning_rate": 1.3449023861171368e-06,
2809
+ "loss": 0.5505,
2810
+ "step": 400
2811
+ },
2812
+ {
2813
+ "epoch": 0.8698481561822126,
2814
+ "grad_norm": 0.7061499953269958,
2815
+ "learning_rate": 1.3232104121475057e-06,
2816
+ "loss": 0.5394,
2817
+ "step": 401
2818
+ },
2819
+ {
2820
+ "epoch": 0.8720173535791758,
2821
+ "grad_norm": 0.5129175186157227,
2822
+ "learning_rate": 1.3015184381778743e-06,
2823
+ "loss": 0.5515,
2824
+ "step": 402
2825
+ },
2826
+ {
2827
+ "epoch": 0.8741865509761388,
2828
+ "grad_norm": 0.6809354424476624,
2829
+ "learning_rate": 1.2798264642082431e-06,
2830
+ "loss": 0.5453,
2831
+ "step": 403
2832
+ },
2833
+ {
2834
+ "epoch": 0.8763557483731019,
2835
+ "grad_norm": 0.5095650553703308,
2836
+ "learning_rate": 1.258134490238612e-06,
2837
+ "loss": 0.5559,
2838
+ "step": 404
2839
+ },
2840
+ {
2841
+ "epoch": 0.8785249457700651,
2842
+ "grad_norm": 0.707577645778656,
2843
+ "learning_rate": 1.2364425162689806e-06,
2844
+ "loss": 0.5318,
2845
+ "step": 405
2846
+ },
2847
+ {
2848
+ "epoch": 0.8806941431670282,
2849
+ "grad_norm": 0.5874403119087219,
2850
+ "learning_rate": 1.2147505422993492e-06,
2851
+ "loss": 0.5481,
2852
+ "step": 406
2853
+ },
2854
+ {
2855
+ "epoch": 0.8828633405639913,
2856
+ "grad_norm": 0.4959114193916321,
2857
+ "learning_rate": 1.193058568329718e-06,
2858
+ "loss": 0.5394,
2859
+ "step": 407
2860
+ },
2861
+ {
2862
+ "epoch": 0.8850325379609545,
2863
+ "grad_norm": 0.5033082365989685,
2864
+ "learning_rate": 1.171366594360087e-06,
2865
+ "loss": 0.5644,
2866
+ "step": 408
2867
+ },
2868
+ {
2869
+ "epoch": 0.8872017353579176,
2870
+ "grad_norm": 0.5417726635932922,
2871
+ "learning_rate": 1.1496746203904555e-06,
2872
+ "loss": 0.5209,
2873
+ "step": 409
2874
+ },
2875
+ {
2876
+ "epoch": 0.8893709327548807,
2877
+ "grad_norm": 0.488228440284729,
2878
+ "learning_rate": 1.1279826464208244e-06,
2879
+ "loss": 0.5406,
2880
+ "step": 410
2881
+ },
2882
+ {
2883
+ "epoch": 0.8915401301518439,
2884
+ "grad_norm": 0.9152227640151978,
2885
+ "learning_rate": 1.1062906724511932e-06,
2886
+ "loss": 0.532,
2887
+ "step": 411
2888
+ },
2889
+ {
2890
+ "epoch": 0.8937093275488069,
2891
+ "grad_norm": 0.5113642811775208,
2892
+ "learning_rate": 1.0845986984815618e-06,
2893
+ "loss": 0.5476,
2894
+ "step": 412
2895
+ },
2896
+ {
2897
+ "epoch": 0.89587852494577,
2898
+ "grad_norm": 0.5312101244926453,
2899
+ "learning_rate": 1.0629067245119307e-06,
2900
+ "loss": 0.5368,
2901
+ "step": 413
2902
+ },
2903
+ {
2904
+ "epoch": 0.8980477223427332,
2905
+ "grad_norm": 0.589953601360321,
2906
+ "learning_rate": 1.0412147505422993e-06,
2907
+ "loss": 0.5381,
2908
+ "step": 414
2909
+ },
2910
+ {
2911
+ "epoch": 0.9002169197396963,
2912
+ "grad_norm": 0.5381547212600708,
2913
+ "learning_rate": 1.0195227765726681e-06,
2914
+ "loss": 0.5492,
2915
+ "step": 415
2916
+ },
2917
+ {
2918
+ "epoch": 0.9023861171366594,
2919
+ "grad_norm": 0.5344692468643188,
2920
+ "learning_rate": 9.97830802603037e-07,
2921
+ "loss": 0.5507,
2922
+ "step": 416
2923
+ },
2924
+ {
2925
+ "epoch": 0.9045553145336226,
2926
+ "grad_norm": 0.48827818036079407,
2927
+ "learning_rate": 9.761388286334058e-07,
2928
+ "loss": 0.5365,
2929
+ "step": 417
2930
+ },
2931
+ {
2932
+ "epoch": 0.9067245119305857,
2933
+ "grad_norm": 0.5503534078598022,
2934
+ "learning_rate": 9.544468546637745e-07,
2935
+ "loss": 0.531,
2936
+ "step": 418
2937
+ },
2938
+ {
2939
+ "epoch": 0.9088937093275488,
2940
+ "grad_norm": 0.5214491486549377,
2941
+ "learning_rate": 9.327548806941432e-07,
2942
+ "loss": 0.5492,
2943
+ "step": 419
2944
+ },
2945
+ {
2946
+ "epoch": 0.911062906724512,
2947
+ "grad_norm": 0.5597646236419678,
2948
+ "learning_rate": 9.11062906724512e-07,
2949
+ "loss": 0.5273,
2950
+ "step": 420
2951
+ },
2952
+ {
2953
+ "epoch": 0.913232104121475,
2954
+ "grad_norm": 0.4814266264438629,
2955
+ "learning_rate": 8.893709327548808e-07,
2956
+ "loss": 0.5467,
2957
+ "step": 421
2958
+ },
2959
+ {
2960
+ "epoch": 0.9154013015184381,
2961
+ "grad_norm": 0.5103883147239685,
2962
+ "learning_rate": 8.676789587852495e-07,
2963
+ "loss": 0.5514,
2964
+ "step": 422
2965
+ },
2966
+ {
2967
+ "epoch": 0.9175704989154013,
2968
+ "grad_norm": 0.4807824492454529,
2969
+ "learning_rate": 8.459869848156182e-07,
2970
+ "loss": 0.5377,
2971
+ "step": 423
2972
+ },
2973
+ {
2974
+ "epoch": 0.9197396963123644,
2975
+ "grad_norm": 0.49137192964553833,
2976
+ "learning_rate": 8.242950108459872e-07,
2977
+ "loss": 0.5384,
2978
+ "step": 424
2979
+ },
2980
+ {
2981
+ "epoch": 0.9219088937093276,
2982
+ "grad_norm": 0.47830575704574585,
2983
+ "learning_rate": 8.026030368763558e-07,
2984
+ "loss": 0.5458,
2985
+ "step": 425
2986
+ },
2987
+ {
2988
+ "epoch": 0.9240780911062907,
2989
+ "grad_norm": 0.5331500768661499,
2990
+ "learning_rate": 7.809110629067245e-07,
2991
+ "loss": 0.514,
2992
+ "step": 426
2993
+ },
2994
+ {
2995
+ "epoch": 0.9262472885032538,
2996
+ "grad_norm": 0.4643237888813019,
2997
+ "learning_rate": 7.592190889370933e-07,
2998
+ "loss": 0.5404,
2999
+ "step": 427
3000
+ },
3001
+ {
3002
+ "epoch": 0.928416485900217,
3003
+ "grad_norm": 0.49474000930786133,
3004
+ "learning_rate": 7.37527114967462e-07,
3005
+ "loss": 0.5687,
3006
+ "step": 428
3007
+ },
3008
+ {
3009
+ "epoch": 0.93058568329718,
3010
+ "grad_norm": 0.551901638507843,
3011
+ "learning_rate": 7.158351409978309e-07,
3012
+ "loss": 0.5485,
3013
+ "step": 429
3014
+ },
3015
+ {
3016
+ "epoch": 0.9327548806941431,
3017
+ "grad_norm": 0.5091038346290588,
3018
+ "learning_rate": 6.941431670281997e-07,
3019
+ "loss": 0.545,
3020
+ "step": 430
3021
+ },
3022
+ {
3023
+ "epoch": 0.9349240780911063,
3024
+ "grad_norm": 0.4894218444824219,
3025
+ "learning_rate": 6.724511930585684e-07,
3026
+ "loss": 0.5391,
3027
+ "step": 431
3028
+ },
3029
+ {
3030
+ "epoch": 0.9370932754880694,
3031
+ "grad_norm": 0.5032333731651306,
3032
+ "learning_rate": 6.507592190889371e-07,
3033
+ "loss": 0.5404,
3034
+ "step": 432
3035
+ },
3036
+ {
3037
+ "epoch": 0.9392624728850325,
3038
+ "grad_norm": 0.5365532040596008,
3039
+ "learning_rate": 6.29067245119306e-07,
3040
+ "loss": 0.524,
3041
+ "step": 433
3042
+ },
3043
+ {
3044
+ "epoch": 0.9414316702819957,
3045
+ "grad_norm": 0.47491738200187683,
3046
+ "learning_rate": 6.073752711496746e-07,
3047
+ "loss": 0.5451,
3048
+ "step": 434
3049
+ },
3050
+ {
3051
+ "epoch": 0.9436008676789588,
3052
+ "grad_norm": 0.4964900314807892,
3053
+ "learning_rate": 5.856832971800435e-07,
3054
+ "loss": 0.5465,
3055
+ "step": 435
3056
+ },
3057
+ {
3058
+ "epoch": 0.9457700650759219,
3059
+ "grad_norm": 0.4745555520057678,
3060
+ "learning_rate": 5.639913232104122e-07,
3061
+ "loss": 0.5432,
3062
+ "step": 436
3063
+ },
3064
+ {
3065
+ "epoch": 0.9479392624728851,
3066
+ "grad_norm": 0.6818556785583496,
3067
+ "learning_rate": 5.422993492407809e-07,
3068
+ "loss": 0.5382,
3069
+ "step": 437
3070
+ },
3071
+ {
3072
+ "epoch": 0.9501084598698482,
3073
+ "grad_norm": 0.5324418544769287,
3074
+ "learning_rate": 5.206073752711497e-07,
3075
+ "loss": 0.5733,
3076
+ "step": 438
3077
+ },
3078
+ {
3079
+ "epoch": 0.9522776572668112,
3080
+ "grad_norm": 0.5001943111419678,
3081
+ "learning_rate": 4.989154013015185e-07,
3082
+ "loss": 0.5331,
3083
+ "step": 439
3084
+ },
3085
+ {
3086
+ "epoch": 0.9544468546637744,
3087
+ "grad_norm": 0.5757445096969604,
3088
+ "learning_rate": 4.772234273318872e-07,
3089
+ "loss": 0.5383,
3090
+ "step": 440
3091
+ },
3092
+ {
3093
+ "epoch": 0.9566160520607375,
3094
+ "grad_norm": 0.6072180271148682,
3095
+ "learning_rate": 4.55531453362256e-07,
3096
+ "loss": 0.5509,
3097
+ "step": 441
3098
+ },
3099
+ {
3100
+ "epoch": 0.9587852494577006,
3101
+ "grad_norm": 0.5225549936294556,
3102
+ "learning_rate": 4.3383947939262475e-07,
3103
+ "loss": 0.5279,
3104
+ "step": 442
3105
+ },
3106
+ {
3107
+ "epoch": 0.9609544468546638,
3108
+ "grad_norm": 0.4668637216091156,
3109
+ "learning_rate": 4.121475054229936e-07,
3110
+ "loss": 0.5488,
3111
+ "step": 443
3112
+ },
3113
+ {
3114
+ "epoch": 0.9631236442516269,
3115
+ "grad_norm": 0.46520882844924927,
3116
+ "learning_rate": 3.9045553145336227e-07,
3117
+ "loss": 0.5272,
3118
+ "step": 444
3119
+ },
3120
+ {
3121
+ "epoch": 0.96529284164859,
3122
+ "grad_norm": 0.45933347940444946,
3123
+ "learning_rate": 3.68763557483731e-07,
3124
+ "loss": 0.5526,
3125
+ "step": 445
3126
+ },
3127
+ {
3128
+ "epoch": 0.9674620390455532,
3129
+ "grad_norm": 0.6567378044128418,
3130
+ "learning_rate": 3.4707158351409984e-07,
3131
+ "loss": 0.5387,
3132
+ "step": 446
3133
+ },
3134
+ {
3135
+ "epoch": 0.9696312364425163,
3136
+ "grad_norm": 0.49450600147247314,
3137
+ "learning_rate": 3.2537960954446857e-07,
3138
+ "loss": 0.5329,
3139
+ "step": 447
3140
+ },
3141
+ {
3142
+ "epoch": 0.9718004338394793,
3143
+ "grad_norm": 0.5677459239959717,
3144
+ "learning_rate": 3.036876355748373e-07,
3145
+ "loss": 0.5459,
3146
+ "step": 448
3147
+ },
3148
+ {
3149
+ "epoch": 0.9739696312364425,
3150
+ "grad_norm": 0.4713030755519867,
3151
+ "learning_rate": 2.819956616052061e-07,
3152
+ "loss": 0.5394,
3153
+ "step": 449
3154
+ },
3155
+ {
3156
+ "epoch": 0.9761388286334056,
3157
+ "grad_norm": 0.5460206866264343,
3158
+ "learning_rate": 2.6030368763557483e-07,
3159
+ "loss": 0.5318,
3160
+ "step": 450
3161
+ },
3162
+ {
3163
+ "epoch": 0.9783080260303688,
3164
+ "grad_norm": 0.4780767560005188,
3165
+ "learning_rate": 2.386117136659436e-07,
3166
+ "loss": 0.5407,
3167
+ "step": 451
3168
+ },
3169
+ {
3170
+ "epoch": 0.9804772234273319,
3171
+ "grad_norm": 0.4803195297718048,
3172
+ "learning_rate": 2.1691973969631237e-07,
3173
+ "loss": 0.5199,
3174
+ "step": 452
3175
+ },
3176
+ {
3177
+ "epoch": 0.982646420824295,
3178
+ "grad_norm": 0.5350447297096252,
3179
+ "learning_rate": 1.9522776572668113e-07,
3180
+ "loss": 0.5434,
3181
+ "step": 453
3182
+ },
3183
+ {
3184
+ "epoch": 0.9848156182212582,
3185
+ "grad_norm": 0.5030990839004517,
3186
+ "learning_rate": 1.7353579175704992e-07,
3187
+ "loss": 0.5368,
3188
+ "step": 454
3189
+ },
3190
+ {
3191
+ "epoch": 0.9869848156182213,
3192
+ "grad_norm": 0.4834834933280945,
3193
+ "learning_rate": 1.5184381778741865e-07,
3194
+ "loss": 0.5386,
3195
+ "step": 455
3196
+ },
3197
+ {
3198
+ "epoch": 0.9891540130151844,
3199
+ "grad_norm": 0.5929272174835205,
3200
+ "learning_rate": 1.3015184381778741e-07,
3201
+ "loss": 0.5452,
3202
+ "step": 456
3203
+ },
3204
+ {
3205
+ "epoch": 0.9913232104121475,
3206
+ "grad_norm": 0.4731636047363281,
3207
+ "learning_rate": 1.0845986984815619e-07,
3208
+ "loss": 0.5294,
3209
+ "step": 457
3210
+ },
3211
+ {
3212
+ "epoch": 0.9934924078091106,
3213
+ "grad_norm": 0.48876383900642395,
3214
+ "learning_rate": 8.676789587852496e-08,
3215
+ "loss": 0.539,
3216
+ "step": 458
3217
+ },
3218
+ {
3219
+ "epoch": 0.9956616052060737,
3220
+ "grad_norm": 0.4793936610221863,
3221
+ "learning_rate": 6.507592190889371e-08,
3222
+ "loss": 0.5496,
3223
+ "step": 459
3224
+ },
3225
+ {
3226
+ "epoch": 0.9978308026030369,
3227
+ "grad_norm": 0.46681129932403564,
3228
+ "learning_rate": 4.338394793926248e-08,
3229
+ "loss": 0.5283,
3230
+ "step": 460
3231
+ },
3232
+ {
3233
+ "epoch": 1.0,
3234
+ "grad_norm": 0.5591140389442444,
3235
+ "learning_rate": 2.169197396963124e-08,
3236
+ "loss": 0.5435,
3237
+ "step": 461
3238
+ }
3239
+ ],
3240
+ "logging_steps": 1.0,
3241
+ "max_steps": 461,
3242
+ "num_input_tokens_seen": 0,
3243
+ "num_train_epochs": 1,
3244
+ "save_steps": 0,
3245
+ "stateful_callbacks": {
3246
+ "TrainerControl": {
3247
+ "args": {
3248
+ "should_epoch_stop": false,
3249
+ "should_evaluate": false,
3250
+ "should_log": false,
3251
+ "should_save": true,
3252
+ "should_training_stop": true
3253
+ },
3254
+ "attributes": {}
3255
+ }
3256
+ },
3257
+ "total_flos": 8.281141529490227e+17,
3258
+ "train_batch_size": 1,
3259
+ "trial_name": null,
3260
+ "trial_params": null
3261
+ }