Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- adapter_config.json +26 -0
- adapter_model.safetensors +3 -0
- checkpoint-2038/adapter_config.json +26 -0
- checkpoint-2038/adapter_model.safetensors +3 -0
- checkpoint-2038/optimizer.pt +3 -0
- checkpoint-2038/rng_state.pth +3 -0
- checkpoint-2038/scheduler.pt +3 -0
- checkpoint-2038/trainer_state.json +1774 -0
- checkpoint-2038/training_args.bin +3 -0
- git_hash.txt +1 -0
- preprocessor_config.json +25 -0
- special_tokens_map.json +39 -0
- tokenizer.json +3 -0
- tokenizer_config.json +0 -0
- training_config.yml +42 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "google/paligemma-3b-ft-ocrvqa-224",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": false,
|
| 8 |
+
"init_lora_weights": "gaussian",
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 32,
|
| 14 |
+
"lora_dropout": 0.1,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 32,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": "(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
|
| 23 |
+
"task_type": "FEATURE_EXTRACTION",
|
| 24 |
+
"use_dora": false,
|
| 25 |
+
"use_rslora": false
|
| 26 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64a4d0d1f5e613119f032fbc4a3b6a0d62aaee2634abf22f6b6ad2bc84859593
|
| 3 |
+
size 78625112
|
checkpoint-2038/adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "google/paligemma-3b-ft-ocrvqa-224",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": false,
|
| 8 |
+
"init_lora_weights": "gaussian",
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 32,
|
| 14 |
+
"lora_dropout": 0.1,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 32,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": "(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
|
| 23 |
+
"task_type": "FEATURE_EXTRACTION",
|
| 24 |
+
"use_dora": false,
|
| 25 |
+
"use_rslora": false
|
| 26 |
+
}
|
checkpoint-2038/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64a4d0d1f5e613119f032fbc4a3b6a0d62aaee2634abf22f6b6ad2bc84859593
|
| 3 |
+
size 78625112
|
checkpoint-2038/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:978ec9ed25c8ea6730d56610c8a2578b214830c406b61921287cc49a444dcba7
|
| 3 |
+
size 157385722
|
checkpoint-2038/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:906c2ebcdd63712eb93ba4d5ca89b84214a092db8695e20fb9dba7a51ee94810
|
| 3 |
+
size 14244
|
checkpoint-2038/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:710f1e838b86ac5bb44884b1d784027122ce88b38e3dac79571ad7a4758c4f9f
|
| 3 |
+
size 1064
|
checkpoint-2038/trainer_state.json
ADDED
|
@@ -0,0 +1,1774 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 2038,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004906771344455349,
|
| 13 |
+
"grad_norm": 0.1728515625,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.6955,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.009813542688910697,
|
| 20 |
+
"grad_norm": 0.291015625,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 0.6952,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.014720314033366046,
|
| 27 |
+
"grad_norm": 0.25,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 0.6964,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.019627085377821395,
|
| 34 |
+
"grad_norm": 0.212890625,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 0.6966,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02453385672227674,
|
| 41 |
+
"grad_norm": 0.302734375,
|
| 42 |
+
"learning_rate": 2.5e-05,
|
| 43 |
+
"loss": 0.6944,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02453385672227674,
|
| 48 |
+
"eval_loss": 0.6937512159347534,
|
| 49 |
+
"eval_runtime": 28.3512,
|
| 50 |
+
"eval_samples_per_second": 7.054,
|
| 51 |
+
"eval_steps_per_second": 1.764,
|
| 52 |
+
"step": 50
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.029440628066732092,
|
| 56 |
+
"grad_norm": 0.2353515625,
|
| 57 |
+
"learning_rate": 3e-05,
|
| 58 |
+
"loss": 0.6944,
|
| 59 |
+
"step": 60
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03434739941118744,
|
| 63 |
+
"grad_norm": 0.14453125,
|
| 64 |
+
"learning_rate": 3.5e-05,
|
| 65 |
+
"loss": 0.694,
|
| 66 |
+
"step": 70
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.03925417075564279,
|
| 70 |
+
"grad_norm": 0.26953125,
|
| 71 |
+
"learning_rate": 4e-05,
|
| 72 |
+
"loss": 0.6938,
|
| 73 |
+
"step": 80
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.04416094210009813,
|
| 77 |
+
"grad_norm": 0.279296875,
|
| 78 |
+
"learning_rate": 4.5e-05,
|
| 79 |
+
"loss": 0.6928,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04906771344455348,
|
| 84 |
+
"grad_norm": 0.2490234375,
|
| 85 |
+
"learning_rate": 5e-05,
|
| 86 |
+
"loss": 0.6928,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04906771344455348,
|
| 91 |
+
"eval_loss": 0.6929062008857727,
|
| 92 |
+
"eval_runtime": 28.6139,
|
| 93 |
+
"eval_samples_per_second": 6.99,
|
| 94 |
+
"eval_steps_per_second": 1.747,
|
| 95 |
+
"step": 100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.053974484789008834,
|
| 99 |
+
"grad_norm": 0.296875,
|
| 100 |
+
"learning_rate": 4.974200206398349e-05,
|
| 101 |
+
"loss": 0.692,
|
| 102 |
+
"step": 110
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.058881256133464184,
|
| 106 |
+
"grad_norm": 0.33984375,
|
| 107 |
+
"learning_rate": 4.948400412796697e-05,
|
| 108 |
+
"loss": 0.6924,
|
| 109 |
+
"step": 120
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.06378802747791953,
|
| 113 |
+
"grad_norm": 0.3359375,
|
| 114 |
+
"learning_rate": 4.922600619195047e-05,
|
| 115 |
+
"loss": 0.6921,
|
| 116 |
+
"step": 130
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.06869479882237488,
|
| 120 |
+
"grad_norm": 0.267578125,
|
| 121 |
+
"learning_rate": 4.896800825593396e-05,
|
| 122 |
+
"loss": 0.6927,
|
| 123 |
+
"step": 140
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.07360157016683022,
|
| 127 |
+
"grad_norm": 0.2236328125,
|
| 128 |
+
"learning_rate": 4.8710010319917446e-05,
|
| 129 |
+
"loss": 0.6914,
|
| 130 |
+
"step": 150
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.07360157016683022,
|
| 134 |
+
"eval_loss": 0.6915245056152344,
|
| 135 |
+
"eval_runtime": 28.6003,
|
| 136 |
+
"eval_samples_per_second": 6.993,
|
| 137 |
+
"eval_steps_per_second": 1.748,
|
| 138 |
+
"step": 150
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.07850834151128558,
|
| 142 |
+
"grad_norm": 0.345703125,
|
| 143 |
+
"learning_rate": 4.845201238390093e-05,
|
| 144 |
+
"loss": 0.6923,
|
| 145 |
+
"step": 160
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.08341511285574092,
|
| 149 |
+
"grad_norm": 0.3046875,
|
| 150 |
+
"learning_rate": 4.819401444788442e-05,
|
| 151 |
+
"loss": 0.6893,
|
| 152 |
+
"step": 170
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.08832188420019627,
|
| 156 |
+
"grad_norm": 0.28125,
|
| 157 |
+
"learning_rate": 4.793601651186791e-05,
|
| 158 |
+
"loss": 0.6865,
|
| 159 |
+
"step": 180
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.09322865554465162,
|
| 163 |
+
"grad_norm": 0.43359375,
|
| 164 |
+
"learning_rate": 4.7678018575851394e-05,
|
| 165 |
+
"loss": 0.6877,
|
| 166 |
+
"step": 190
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.09813542688910697,
|
| 170 |
+
"grad_norm": 0.42578125,
|
| 171 |
+
"learning_rate": 4.742002063983488e-05,
|
| 172 |
+
"loss": 0.6852,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.09813542688910697,
|
| 177 |
+
"eval_loss": 0.6849114894866943,
|
| 178 |
+
"eval_runtime": 28.416,
|
| 179 |
+
"eval_samples_per_second": 7.038,
|
| 180 |
+
"eval_steps_per_second": 1.76,
|
| 181 |
+
"step": 200
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.10304219823356231,
|
| 185 |
+
"grad_norm": 0.39453125,
|
| 186 |
+
"learning_rate": 4.716202270381837e-05,
|
| 187 |
+
"loss": 0.6767,
|
| 188 |
+
"step": 210
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.10794896957801767,
|
| 192 |
+
"grad_norm": 0.45703125,
|
| 193 |
+
"learning_rate": 4.690402476780186e-05,
|
| 194 |
+
"loss": 0.6804,
|
| 195 |
+
"step": 220
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.11285574092247301,
|
| 199 |
+
"grad_norm": 0.78125,
|
| 200 |
+
"learning_rate": 4.664602683178535e-05,
|
| 201 |
+
"loss": 0.6814,
|
| 202 |
+
"step": 230
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.11776251226692837,
|
| 206 |
+
"grad_norm": 2.046875,
|
| 207 |
+
"learning_rate": 4.638802889576884e-05,
|
| 208 |
+
"loss": 0.6586,
|
| 209 |
+
"step": 240
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.12266928361138371,
|
| 213 |
+
"grad_norm": 1.828125,
|
| 214 |
+
"learning_rate": 4.613003095975233e-05,
|
| 215 |
+
"loss": 0.6478,
|
| 216 |
+
"step": 250
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.12266928361138371,
|
| 220 |
+
"eval_loss": 0.6532555222511292,
|
| 221 |
+
"eval_runtime": 28.4259,
|
| 222 |
+
"eval_samples_per_second": 7.036,
|
| 223 |
+
"eval_steps_per_second": 1.759,
|
| 224 |
+
"step": 250
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.12757605495583907,
|
| 228 |
+
"grad_norm": 1.90625,
|
| 229 |
+
"learning_rate": 4.587203302373581e-05,
|
| 230 |
+
"loss": 0.6473,
|
| 231 |
+
"step": 260
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.1324828263002944,
|
| 235 |
+
"grad_norm": 1.84375,
|
| 236 |
+
"learning_rate": 4.56140350877193e-05,
|
| 237 |
+
"loss": 0.6397,
|
| 238 |
+
"step": 270
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.13738959764474976,
|
| 242 |
+
"grad_norm": 2.046875,
|
| 243 |
+
"learning_rate": 4.535603715170279e-05,
|
| 244 |
+
"loss": 0.6243,
|
| 245 |
+
"step": 280
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.1422963689892051,
|
| 249 |
+
"grad_norm": 1.640625,
|
| 250 |
+
"learning_rate": 4.5098039215686275e-05,
|
| 251 |
+
"loss": 0.6304,
|
| 252 |
+
"step": 290
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.14720314033366044,
|
| 256 |
+
"grad_norm": 1.84375,
|
| 257 |
+
"learning_rate": 4.4840041279669764e-05,
|
| 258 |
+
"loss": 0.6278,
|
| 259 |
+
"step": 300
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.14720314033366044,
|
| 263 |
+
"eval_loss": 0.6153883337974548,
|
| 264 |
+
"eval_runtime": 28.6968,
|
| 265 |
+
"eval_samples_per_second": 6.969,
|
| 266 |
+
"eval_steps_per_second": 1.742,
|
| 267 |
+
"step": 300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.1521099116781158,
|
| 271 |
+
"grad_norm": 1.5,
|
| 272 |
+
"learning_rate": 4.458204334365325e-05,
|
| 273 |
+
"loss": 0.6067,
|
| 274 |
+
"step": 310
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.15701668302257116,
|
| 278 |
+
"grad_norm": 1.7734375,
|
| 279 |
+
"learning_rate": 4.432404540763674e-05,
|
| 280 |
+
"loss": 0.5956,
|
| 281 |
+
"step": 320
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.1619234543670265,
|
| 285 |
+
"grad_norm": 1.640625,
|
| 286 |
+
"learning_rate": 4.406604747162023e-05,
|
| 287 |
+
"loss": 0.6263,
|
| 288 |
+
"step": 330
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.16683022571148184,
|
| 292 |
+
"grad_norm": 1.5,
|
| 293 |
+
"learning_rate": 4.380804953560372e-05,
|
| 294 |
+
"loss": 0.5947,
|
| 295 |
+
"step": 340
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.1717369970559372,
|
| 299 |
+
"grad_norm": 1.0390625,
|
| 300 |
+
"learning_rate": 4.355005159958721e-05,
|
| 301 |
+
"loss": 0.609,
|
| 302 |
+
"step": 350
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.1717369970559372,
|
| 306 |
+
"eval_loss": 0.6010438799858093,
|
| 307 |
+
"eval_runtime": 28.6965,
|
| 308 |
+
"eval_samples_per_second": 6.969,
|
| 309 |
+
"eval_steps_per_second": 1.742,
|
| 310 |
+
"step": 350
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.17664376840039253,
|
| 314 |
+
"grad_norm": 1.90625,
|
| 315 |
+
"learning_rate": 4.329205366357069e-05,
|
| 316 |
+
"loss": 0.5948,
|
| 317 |
+
"step": 360
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1815505397448479,
|
| 321 |
+
"grad_norm": 1.546875,
|
| 322 |
+
"learning_rate": 4.303405572755418e-05,
|
| 323 |
+
"loss": 0.6043,
|
| 324 |
+
"step": 370
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.18645731108930325,
|
| 328 |
+
"grad_norm": 1.71875,
|
| 329 |
+
"learning_rate": 4.2776057791537674e-05,
|
| 330 |
+
"loss": 0.5791,
|
| 331 |
+
"step": 380
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.19136408243375858,
|
| 335 |
+
"grad_norm": 1.2109375,
|
| 336 |
+
"learning_rate": 4.2518059855521156e-05,
|
| 337 |
+
"loss": 0.56,
|
| 338 |
+
"step": 390
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.19627085377821393,
|
| 342 |
+
"grad_norm": 1.7734375,
|
| 343 |
+
"learning_rate": 4.2260061919504645e-05,
|
| 344 |
+
"loss": 0.5801,
|
| 345 |
+
"step": 400
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.19627085377821393,
|
| 349 |
+
"eval_loss": 0.5847232937812805,
|
| 350 |
+
"eval_runtime": 28.6639,
|
| 351 |
+
"eval_samples_per_second": 6.977,
|
| 352 |
+
"eval_steps_per_second": 1.744,
|
| 353 |
+
"step": 400
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.2011776251226693,
|
| 357 |
+
"grad_norm": 1.703125,
|
| 358 |
+
"learning_rate": 4.200206398348813e-05,
|
| 359 |
+
"loss": 0.5517,
|
| 360 |
+
"step": 410
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.20608439646712462,
|
| 364 |
+
"grad_norm": 1.296875,
|
| 365 |
+
"learning_rate": 4.174406604747162e-05,
|
| 366 |
+
"loss": 0.5799,
|
| 367 |
+
"step": 420
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.21099116781157998,
|
| 371 |
+
"grad_norm": 1.1953125,
|
| 372 |
+
"learning_rate": 4.148606811145511e-05,
|
| 373 |
+
"loss": 0.54,
|
| 374 |
+
"step": 430
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.21589793915603533,
|
| 378 |
+
"grad_norm": 3.859375,
|
| 379 |
+
"learning_rate": 4.12280701754386e-05,
|
| 380 |
+
"loss": 0.5591,
|
| 381 |
+
"step": 440
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.22080471050049066,
|
| 385 |
+
"grad_norm": 1.1484375,
|
| 386 |
+
"learning_rate": 4.097007223942209e-05,
|
| 387 |
+
"loss": 0.5558,
|
| 388 |
+
"step": 450
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.22080471050049066,
|
| 392 |
+
"eval_loss": 0.5724604725837708,
|
| 393 |
+
"eval_runtime": 28.6438,
|
| 394 |
+
"eval_samples_per_second": 6.982,
|
| 395 |
+
"eval_steps_per_second": 1.746,
|
| 396 |
+
"step": 450
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.22571148184494602,
|
| 400 |
+
"grad_norm": 1.2265625,
|
| 401 |
+
"learning_rate": 4.071207430340557e-05,
|
| 402 |
+
"loss": 0.6203,
|
| 403 |
+
"step": 460
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.23061825318940138,
|
| 407 |
+
"grad_norm": 1.5234375,
|
| 408 |
+
"learning_rate": 4.0454076367389066e-05,
|
| 409 |
+
"loss": 0.5813,
|
| 410 |
+
"step": 470
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.23552502453385674,
|
| 414 |
+
"grad_norm": 2.5,
|
| 415 |
+
"learning_rate": 4.0196078431372555e-05,
|
| 416 |
+
"loss": 0.5946,
|
| 417 |
+
"step": 480
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.24043179587831207,
|
| 421 |
+
"grad_norm": 1.671875,
|
| 422 |
+
"learning_rate": 3.9938080495356037e-05,
|
| 423 |
+
"loss": 0.5635,
|
| 424 |
+
"step": 490
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.24533856722276742,
|
| 428 |
+
"grad_norm": 1.3125,
|
| 429 |
+
"learning_rate": 3.9680082559339525e-05,
|
| 430 |
+
"loss": 0.5625,
|
| 431 |
+
"step": 500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.24533856722276742,
|
| 435 |
+
"eval_loss": 0.5732089877128601,
|
| 436 |
+
"eval_runtime": 28.6521,
|
| 437 |
+
"eval_samples_per_second": 6.98,
|
| 438 |
+
"eval_steps_per_second": 1.745,
|
| 439 |
+
"step": 500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.25024533856722275,
|
| 443 |
+
"grad_norm": 1.453125,
|
| 444 |
+
"learning_rate": 3.9422084623323014e-05,
|
| 445 |
+
"loss": 0.5378,
|
| 446 |
+
"step": 510
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.25515210991167814,
|
| 450 |
+
"grad_norm": 1.046875,
|
| 451 |
+
"learning_rate": 3.91640866873065e-05,
|
| 452 |
+
"loss": 0.5908,
|
| 453 |
+
"step": 520
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.26005888125613347,
|
| 457 |
+
"grad_norm": 1.2421875,
|
| 458 |
+
"learning_rate": 3.890608875128999e-05,
|
| 459 |
+
"loss": 0.576,
|
| 460 |
+
"step": 530
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.2649656526005888,
|
| 464 |
+
"grad_norm": 0.87109375,
|
| 465 |
+
"learning_rate": 3.864809081527348e-05,
|
| 466 |
+
"loss": 0.5248,
|
| 467 |
+
"step": 540
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.2698724239450442,
|
| 471 |
+
"grad_norm": 2.046875,
|
| 472 |
+
"learning_rate": 3.839009287925697e-05,
|
| 473 |
+
"loss": 0.5448,
|
| 474 |
+
"step": 550
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.2698724239450442,
|
| 478 |
+
"eval_loss": 0.5607851147651672,
|
| 479 |
+
"eval_runtime": 28.5118,
|
| 480 |
+
"eval_samples_per_second": 7.015,
|
| 481 |
+
"eval_steps_per_second": 1.754,
|
| 482 |
+
"step": 550
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.2747791952894995,
|
| 486 |
+
"grad_norm": 1.828125,
|
| 487 |
+
"learning_rate": 3.813209494324045e-05,
|
| 488 |
+
"loss": 0.5693,
|
| 489 |
+
"step": 560
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.27968596663395484,
|
| 493 |
+
"grad_norm": 0.65234375,
|
| 494 |
+
"learning_rate": 3.7874097007223946e-05,
|
| 495 |
+
"loss": 0.5248,
|
| 496 |
+
"step": 570
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.2845927379784102,
|
| 500 |
+
"grad_norm": 1.4453125,
|
| 501 |
+
"learning_rate": 3.7616099071207435e-05,
|
| 502 |
+
"loss": 0.5815,
|
| 503 |
+
"step": 580
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 0.28949950932286556,
|
| 507 |
+
"grad_norm": 1.0078125,
|
| 508 |
+
"learning_rate": 3.735810113519092e-05,
|
| 509 |
+
"loss": 0.5524,
|
| 510 |
+
"step": 590
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.2944062806673209,
|
| 514 |
+
"grad_norm": 1.3984375,
|
| 515 |
+
"learning_rate": 3.7100103199174406e-05,
|
| 516 |
+
"loss": 0.5517,
|
| 517 |
+
"step": 600
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 0.2944062806673209,
|
| 521 |
+
"eval_loss": 0.5586492419242859,
|
| 522 |
+
"eval_runtime": 28.5258,
|
| 523 |
+
"eval_samples_per_second": 7.011,
|
| 524 |
+
"eval_steps_per_second": 1.753,
|
| 525 |
+
"step": 600
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.29931305201177627,
|
| 529 |
+
"grad_norm": 1.375,
|
| 530 |
+
"learning_rate": 3.6842105263157895e-05,
|
| 531 |
+
"loss": 0.5308,
|
| 532 |
+
"step": 610
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.3042198233562316,
|
| 536 |
+
"grad_norm": 1.3046875,
|
| 537 |
+
"learning_rate": 3.658410732714139e-05,
|
| 538 |
+
"loss": 0.5938,
|
| 539 |
+
"step": 620
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.30912659470068693,
|
| 543 |
+
"grad_norm": 1.78125,
|
| 544 |
+
"learning_rate": 3.632610939112487e-05,
|
| 545 |
+
"loss": 0.5611,
|
| 546 |
+
"step": 630
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.3140333660451423,
|
| 550 |
+
"grad_norm": 0.74609375,
|
| 551 |
+
"learning_rate": 3.606811145510836e-05,
|
| 552 |
+
"loss": 0.5709,
|
| 553 |
+
"step": 640
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.31894013738959764,
|
| 557 |
+
"grad_norm": 0.99609375,
|
| 558 |
+
"learning_rate": 3.581011351909185e-05,
|
| 559 |
+
"loss": 0.6033,
|
| 560 |
+
"step": 650
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.31894013738959764,
|
| 564 |
+
"eval_loss": 0.564706027507782,
|
| 565 |
+
"eval_runtime": 28.5712,
|
| 566 |
+
"eval_samples_per_second": 7.0,
|
| 567 |
+
"eval_steps_per_second": 1.75,
|
| 568 |
+
"step": 650
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.323846908734053,
|
| 572 |
+
"grad_norm": 1.2578125,
|
| 573 |
+
"learning_rate": 3.555211558307533e-05,
|
| 574 |
+
"loss": 0.5123,
|
| 575 |
+
"step": 660
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.32875368007850836,
|
| 579 |
+
"grad_norm": 0.97265625,
|
| 580 |
+
"learning_rate": 3.529411764705883e-05,
|
| 581 |
+
"loss": 0.5205,
|
| 582 |
+
"step": 670
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 0.3336604514229637,
|
| 586 |
+
"grad_norm": 1.6953125,
|
| 587 |
+
"learning_rate": 3.5036119711042316e-05,
|
| 588 |
+
"loss": 0.5854,
|
| 589 |
+
"step": 680
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.338567222767419,
|
| 593 |
+
"grad_norm": 1.453125,
|
| 594 |
+
"learning_rate": 3.4778121775025805e-05,
|
| 595 |
+
"loss": 0.5375,
|
| 596 |
+
"step": 690
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.3434739941118744,
|
| 600 |
+
"grad_norm": 1.03125,
|
| 601 |
+
"learning_rate": 3.452012383900929e-05,
|
| 602 |
+
"loss": 0.5336,
|
| 603 |
+
"step": 700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.3434739941118744,
|
| 607 |
+
"eval_loss": 0.5557608008384705,
|
| 608 |
+
"eval_runtime": 28.4609,
|
| 609 |
+
"eval_samples_per_second": 7.027,
|
| 610 |
+
"eval_steps_per_second": 1.757,
|
| 611 |
+
"step": 700
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.34838076545632973,
|
| 615 |
+
"grad_norm": 1.5859375,
|
| 616 |
+
"learning_rate": 3.4262125902992775e-05,
|
| 617 |
+
"loss": 0.5533,
|
| 618 |
+
"step": 710
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.35328753680078506,
|
| 622 |
+
"grad_norm": 0.8984375,
|
| 623 |
+
"learning_rate": 3.400412796697627e-05,
|
| 624 |
+
"loss": 0.5333,
|
| 625 |
+
"step": 720
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.35819430814524045,
|
| 629 |
+
"grad_norm": 1.09375,
|
| 630 |
+
"learning_rate": 3.374613003095975e-05,
|
| 631 |
+
"loss": 0.5536,
|
| 632 |
+
"step": 730
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3631010794896958,
|
| 636 |
+
"grad_norm": 1.1171875,
|
| 637 |
+
"learning_rate": 3.348813209494324e-05,
|
| 638 |
+
"loss": 0.5448,
|
| 639 |
+
"step": 740
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3680078508341511,
|
| 643 |
+
"grad_norm": 0.96875,
|
| 644 |
+
"learning_rate": 3.323013415892673e-05,
|
| 645 |
+
"loss": 0.532,
|
| 646 |
+
"step": 750
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3680078508341511,
|
| 650 |
+
"eval_loss": 0.5521541833877563,
|
| 651 |
+
"eval_runtime": 28.4595,
|
| 652 |
+
"eval_samples_per_second": 7.028,
|
| 653 |
+
"eval_steps_per_second": 1.757,
|
| 654 |
+
"step": 750
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.3729146221786065,
|
| 658 |
+
"grad_norm": 0.90625,
|
| 659 |
+
"learning_rate": 3.297213622291022e-05,
|
| 660 |
+
"loss": 0.565,
|
| 661 |
+
"step": 760
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.3778213935230618,
|
| 665 |
+
"grad_norm": 1.4140625,
|
| 666 |
+
"learning_rate": 3.271413828689371e-05,
|
| 667 |
+
"loss": 0.5278,
|
| 668 |
+
"step": 770
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.38272816486751715,
|
| 672 |
+
"grad_norm": 1.3359375,
|
| 673 |
+
"learning_rate": 3.24561403508772e-05,
|
| 674 |
+
"loss": 0.5582,
|
| 675 |
+
"step": 780
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.38763493621197254,
|
| 679 |
+
"grad_norm": 0.84375,
|
| 680 |
+
"learning_rate": 3.2198142414860685e-05,
|
| 681 |
+
"loss": 0.5253,
|
| 682 |
+
"step": 790
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.39254170755642787,
|
| 686 |
+
"grad_norm": 1.4296875,
|
| 687 |
+
"learning_rate": 3.194014447884417e-05,
|
| 688 |
+
"loss": 0.5596,
|
| 689 |
+
"step": 800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.39254170755642787,
|
| 693 |
+
"eval_loss": 0.551950216293335,
|
| 694 |
+
"eval_runtime": 28.6541,
|
| 695 |
+
"eval_samples_per_second": 6.98,
|
| 696 |
+
"eval_steps_per_second": 1.745,
|
| 697 |
+
"step": 800
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.3974484789008832,
|
| 701 |
+
"grad_norm": 1.484375,
|
| 702 |
+
"learning_rate": 3.1682146542827656e-05,
|
| 703 |
+
"loss": 0.5513,
|
| 704 |
+
"step": 810
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.4023552502453386,
|
| 708 |
+
"grad_norm": 1.21875,
|
| 709 |
+
"learning_rate": 3.142414860681115e-05,
|
| 710 |
+
"loss": 0.5023,
|
| 711 |
+
"step": 820
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.4072620215897939,
|
| 715 |
+
"grad_norm": 1.109375,
|
| 716 |
+
"learning_rate": 3.1166150670794634e-05,
|
| 717 |
+
"loss": 0.5519,
|
| 718 |
+
"step": 830
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.41216879293424924,
|
| 722 |
+
"grad_norm": 1.3984375,
|
| 723 |
+
"learning_rate": 3.090815273477812e-05,
|
| 724 |
+
"loss": 0.51,
|
| 725 |
+
"step": 840
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.4170755642787046,
|
| 729 |
+
"grad_norm": 1.0859375,
|
| 730 |
+
"learning_rate": 3.065015479876161e-05,
|
| 731 |
+
"loss": 0.5496,
|
| 732 |
+
"step": 850
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.4170755642787046,
|
| 736 |
+
"eval_loss": 0.5526400208473206,
|
| 737 |
+
"eval_runtime": 28.572,
|
| 738 |
+
"eval_samples_per_second": 7.0,
|
| 739 |
+
"eval_steps_per_second": 1.75,
|
| 740 |
+
"step": 850
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.42198233562315995,
|
| 744 |
+
"grad_norm": 1.2265625,
|
| 745 |
+
"learning_rate": 3.0392156862745097e-05,
|
| 746 |
+
"loss": 0.5372,
|
| 747 |
+
"step": 860
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.4268891069676153,
|
| 751 |
+
"grad_norm": 1.9140625,
|
| 752 |
+
"learning_rate": 3.013415892672859e-05,
|
| 753 |
+
"loss": 0.5384,
|
| 754 |
+
"step": 870
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.43179587831207067,
|
| 758 |
+
"grad_norm": 0.99609375,
|
| 759 |
+
"learning_rate": 2.9876160990712077e-05,
|
| 760 |
+
"loss": 0.4739,
|
| 761 |
+
"step": 880
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.436702649656526,
|
| 765 |
+
"grad_norm": 1.15625,
|
| 766 |
+
"learning_rate": 2.9618163054695563e-05,
|
| 767 |
+
"loss": 0.4766,
|
| 768 |
+
"step": 890
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.44160942100098133,
|
| 772 |
+
"grad_norm": 1.7734375,
|
| 773 |
+
"learning_rate": 2.936016511867905e-05,
|
| 774 |
+
"loss": 0.5321,
|
| 775 |
+
"step": 900
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.44160942100098133,
|
| 779 |
+
"eval_loss": 0.5481389164924622,
|
| 780 |
+
"eval_runtime": 28.6453,
|
| 781 |
+
"eval_samples_per_second": 6.982,
|
| 782 |
+
"eval_steps_per_second": 1.745,
|
| 783 |
+
"step": 900
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.4465161923454367,
|
| 787 |
+
"grad_norm": 0.9453125,
|
| 788 |
+
"learning_rate": 2.9102167182662537e-05,
|
| 789 |
+
"loss": 0.5654,
|
| 790 |
+
"step": 910
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.45142296368989204,
|
| 794 |
+
"grad_norm": 1.09375,
|
| 795 |
+
"learning_rate": 2.884416924664603e-05,
|
| 796 |
+
"loss": 0.5514,
|
| 797 |
+
"step": 920
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.4563297350343474,
|
| 801 |
+
"grad_norm": 1.1875,
|
| 802 |
+
"learning_rate": 2.8586171310629518e-05,
|
| 803 |
+
"loss": 0.4833,
|
| 804 |
+
"step": 930
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.46123650637880276,
|
| 808 |
+
"grad_norm": 1.4609375,
|
| 809 |
+
"learning_rate": 2.8328173374613003e-05,
|
| 810 |
+
"loss": 0.549,
|
| 811 |
+
"step": 940
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.4661432777232581,
|
| 815 |
+
"grad_norm": 0.9453125,
|
| 816 |
+
"learning_rate": 2.8070175438596492e-05,
|
| 817 |
+
"loss": 0.5194,
|
| 818 |
+
"step": 950
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.4661432777232581,
|
| 822 |
+
"eval_loss": 0.5417376160621643,
|
| 823 |
+
"eval_runtime": 28.3949,
|
| 824 |
+
"eval_samples_per_second": 7.044,
|
| 825 |
+
"eval_steps_per_second": 1.761,
|
| 826 |
+
"step": 950
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.47105004906771347,
|
| 830 |
+
"grad_norm": 1.4296875,
|
| 831 |
+
"learning_rate": 2.7812177502579977e-05,
|
| 832 |
+
"loss": 0.5468,
|
| 833 |
+
"step": 960
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.4759568204121688,
|
| 837 |
+
"grad_norm": 0.859375,
|
| 838 |
+
"learning_rate": 2.755417956656347e-05,
|
| 839 |
+
"loss": 0.5307,
|
| 840 |
+
"step": 970
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.48086359175662413,
|
| 844 |
+
"grad_norm": 1.6640625,
|
| 845 |
+
"learning_rate": 2.7296181630546958e-05,
|
| 846 |
+
"loss": 0.5379,
|
| 847 |
+
"step": 980
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 0.4857703631010795,
|
| 851 |
+
"grad_norm": 1.8828125,
|
| 852 |
+
"learning_rate": 2.7038183694530443e-05,
|
| 853 |
+
"loss": 0.5197,
|
| 854 |
+
"step": 990
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.49067713444553485,
|
| 858 |
+
"grad_norm": 1.265625,
|
| 859 |
+
"learning_rate": 2.6780185758513932e-05,
|
| 860 |
+
"loss": 0.5557,
|
| 861 |
+
"step": 1000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 0.49067713444553485,
|
| 865 |
+
"eval_loss": 0.5419167876243591,
|
| 866 |
+
"eval_runtime": 28.5704,
|
| 867 |
+
"eval_samples_per_second": 7.0,
|
| 868 |
+
"eval_steps_per_second": 1.75,
|
| 869 |
+
"step": 1000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.4955839057899902,
|
| 873 |
+
"grad_norm": 0.984375,
|
| 874 |
+
"learning_rate": 2.6522187822497424e-05,
|
| 875 |
+
"loss": 0.5121,
|
| 876 |
+
"step": 1010
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 0.5004906771344455,
|
| 880 |
+
"grad_norm": 1.7421875,
|
| 881 |
+
"learning_rate": 2.626418988648091e-05,
|
| 882 |
+
"loss": 0.4932,
|
| 883 |
+
"step": 1020
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.5053974484789009,
|
| 887 |
+
"grad_norm": 1.1015625,
|
| 888 |
+
"learning_rate": 2.60061919504644e-05,
|
| 889 |
+
"loss": 0.5361,
|
| 890 |
+
"step": 1030
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.5103042198233563,
|
| 894 |
+
"grad_norm": 1.1015625,
|
| 895 |
+
"learning_rate": 2.5748194014447884e-05,
|
| 896 |
+
"loss": 0.5107,
|
| 897 |
+
"step": 1040
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.5152109911678115,
|
| 901 |
+
"grad_norm": 1.359375,
|
| 902 |
+
"learning_rate": 2.5490196078431373e-05,
|
| 903 |
+
"loss": 0.5111,
|
| 904 |
+
"step": 1050
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.5152109911678115,
|
| 908 |
+
"eval_loss": 0.5413140654563904,
|
| 909 |
+
"eval_runtime": 28.49,
|
| 910 |
+
"eval_samples_per_second": 7.02,
|
| 911 |
+
"eval_steps_per_second": 1.755,
|
| 912 |
+
"step": 1050
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.5201177625122669,
|
| 916 |
+
"grad_norm": 0.57421875,
|
| 917 |
+
"learning_rate": 2.5232198142414865e-05,
|
| 918 |
+
"loss": 0.5302,
|
| 919 |
+
"step": 1060
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5250245338567223,
|
| 923 |
+
"grad_norm": 1.859375,
|
| 924 |
+
"learning_rate": 2.497420020639835e-05,
|
| 925 |
+
"loss": 0.5089,
|
| 926 |
+
"step": 1070
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5299313052011776,
|
| 930 |
+
"grad_norm": 1.78125,
|
| 931 |
+
"learning_rate": 2.471620227038184e-05,
|
| 932 |
+
"loss": 0.5451,
|
| 933 |
+
"step": 1080
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.534838076545633,
|
| 937 |
+
"grad_norm": 1.0546875,
|
| 938 |
+
"learning_rate": 2.4458204334365324e-05,
|
| 939 |
+
"loss": 0.5736,
|
| 940 |
+
"step": 1090
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5397448478900884,
|
| 944 |
+
"grad_norm": 1.28125,
|
| 945 |
+
"learning_rate": 2.4200206398348816e-05,
|
| 946 |
+
"loss": 0.5065,
|
| 947 |
+
"step": 1100
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5397448478900884,
|
| 951 |
+
"eval_loss": 0.5397326350212097,
|
| 952 |
+
"eval_runtime": 28.806,
|
| 953 |
+
"eval_samples_per_second": 6.943,
|
| 954 |
+
"eval_steps_per_second": 1.736,
|
| 955 |
+
"step": 1100
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5446516192345436,
|
| 959 |
+
"grad_norm": 1.046875,
|
| 960 |
+
"learning_rate": 2.39422084623323e-05,
|
| 961 |
+
"loss": 0.5719,
|
| 962 |
+
"step": 1110
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.549558390578999,
|
| 966 |
+
"grad_norm": 0.75390625,
|
| 967 |
+
"learning_rate": 2.368421052631579e-05,
|
| 968 |
+
"loss": 0.5343,
|
| 969 |
+
"step": 1120
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5544651619234544,
|
| 973 |
+
"grad_norm": 1.6484375,
|
| 974 |
+
"learning_rate": 2.342621259029928e-05,
|
| 975 |
+
"loss": 0.523,
|
| 976 |
+
"step": 1130
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5593719332679097,
|
| 980 |
+
"grad_norm": 0.9765625,
|
| 981 |
+
"learning_rate": 2.3168214654282765e-05,
|
| 982 |
+
"loss": 0.5345,
|
| 983 |
+
"step": 1140
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5642787046123651,
|
| 987 |
+
"grad_norm": 0.8359375,
|
| 988 |
+
"learning_rate": 2.2910216718266257e-05,
|
| 989 |
+
"loss": 0.5264,
|
| 990 |
+
"step": 1150
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5642787046123651,
|
| 994 |
+
"eval_loss": 0.5375524759292603,
|
| 995 |
+
"eval_runtime": 28.5508,
|
| 996 |
+
"eval_samples_per_second": 7.005,
|
| 997 |
+
"eval_steps_per_second": 1.751,
|
| 998 |
+
"step": 1150
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.5691854759568205,
|
| 1002 |
+
"grad_norm": 0.9609375,
|
| 1003 |
+
"learning_rate": 2.2652218782249742e-05,
|
| 1004 |
+
"loss": 0.4863,
|
| 1005 |
+
"step": 1160
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.5740922473012757,
|
| 1009 |
+
"grad_norm": 1.6875,
|
| 1010 |
+
"learning_rate": 2.2394220846233234e-05,
|
| 1011 |
+
"loss": 0.4676,
|
| 1012 |
+
"step": 1170
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.5789990186457311,
|
| 1016 |
+
"grad_norm": 1.3984375,
|
| 1017 |
+
"learning_rate": 2.213622291021672e-05,
|
| 1018 |
+
"loss": 0.5751,
|
| 1019 |
+
"step": 1180
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 0.5839057899901865,
|
| 1023 |
+
"grad_norm": 2.15625,
|
| 1024 |
+
"learning_rate": 2.1878224974200205e-05,
|
| 1025 |
+
"loss": 0.5694,
|
| 1026 |
+
"step": 1190
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 0.5888125613346418,
|
| 1030 |
+
"grad_norm": 0.98828125,
|
| 1031 |
+
"learning_rate": 2.1620227038183697e-05,
|
| 1032 |
+
"loss": 0.5529,
|
| 1033 |
+
"step": 1200
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.5888125613346418,
|
| 1037 |
+
"eval_loss": 0.5384119153022766,
|
| 1038 |
+
"eval_runtime": 28.624,
|
| 1039 |
+
"eval_samples_per_second": 6.987,
|
| 1040 |
+
"eval_steps_per_second": 1.747,
|
| 1041 |
+
"step": 1200
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.5937193326790972,
|
| 1045 |
+
"grad_norm": 1.984375,
|
| 1046 |
+
"learning_rate": 2.1362229102167182e-05,
|
| 1047 |
+
"loss": 0.5438,
|
| 1048 |
+
"step": 1210
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 0.5986261040235525,
|
| 1052 |
+
"grad_norm": 1.3203125,
|
| 1053 |
+
"learning_rate": 2.1104231166150675e-05,
|
| 1054 |
+
"loss": 0.5174,
|
| 1055 |
+
"step": 1220
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 0.6035328753680078,
|
| 1059 |
+
"grad_norm": 1.5078125,
|
| 1060 |
+
"learning_rate": 2.084623323013416e-05,
|
| 1061 |
+
"loss": 0.5421,
|
| 1062 |
+
"step": 1230
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 0.6084396467124632,
|
| 1066 |
+
"grad_norm": 1.234375,
|
| 1067 |
+
"learning_rate": 2.058823529411765e-05,
|
| 1068 |
+
"loss": 0.5427,
|
| 1069 |
+
"step": 1240
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 0.6133464180569186,
|
| 1073 |
+
"grad_norm": 0.95703125,
|
| 1074 |
+
"learning_rate": 2.0330237358101137e-05,
|
| 1075 |
+
"loss": 0.5356,
|
| 1076 |
+
"step": 1250
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 0.6133464180569186,
|
| 1080 |
+
"eval_loss": 0.5382110476493835,
|
| 1081 |
+
"eval_runtime": 28.6288,
|
| 1082 |
+
"eval_samples_per_second": 6.986,
|
| 1083 |
+
"eval_steps_per_second": 1.746,
|
| 1084 |
+
"step": 1250
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 0.6182531894013739,
|
| 1088 |
+
"grad_norm": 1.53125,
|
| 1089 |
+
"learning_rate": 2.0072239422084623e-05,
|
| 1090 |
+
"loss": 0.5667,
|
| 1091 |
+
"step": 1260
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.6231599607458292,
|
| 1095 |
+
"grad_norm": 1.5078125,
|
| 1096 |
+
"learning_rate": 1.9814241486068115e-05,
|
| 1097 |
+
"loss": 0.5421,
|
| 1098 |
+
"step": 1270
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 0.6280667320902846,
|
| 1102 |
+
"grad_norm": 1.6015625,
|
| 1103 |
+
"learning_rate": 1.95562435500516e-05,
|
| 1104 |
+
"loss": 0.4955,
|
| 1105 |
+
"step": 1280
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.6329735034347399,
|
| 1109 |
+
"grad_norm": 1.2890625,
|
| 1110 |
+
"learning_rate": 1.929824561403509e-05,
|
| 1111 |
+
"loss": 0.4995,
|
| 1112 |
+
"step": 1290
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.6378802747791953,
|
| 1116 |
+
"grad_norm": 1.4140625,
|
| 1117 |
+
"learning_rate": 1.9040247678018578e-05,
|
| 1118 |
+
"loss": 0.5029,
|
| 1119 |
+
"step": 1300
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.6378802747791953,
|
| 1123 |
+
"eval_loss": 0.5337280035018921,
|
| 1124 |
+
"eval_runtime": 28.4662,
|
| 1125 |
+
"eval_samples_per_second": 7.026,
|
| 1126 |
+
"eval_steps_per_second": 1.756,
|
| 1127 |
+
"step": 1300
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 0.6427870461236507,
|
| 1131 |
+
"grad_norm": 1.3203125,
|
| 1132 |
+
"learning_rate": 1.8782249742002063e-05,
|
| 1133 |
+
"loss": 0.5149,
|
| 1134 |
+
"step": 1310
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 0.647693817468106,
|
| 1138 |
+
"grad_norm": 0.91796875,
|
| 1139 |
+
"learning_rate": 1.8524251805985555e-05,
|
| 1140 |
+
"loss": 0.5011,
|
| 1141 |
+
"step": 1320
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 0.6526005888125613,
|
| 1145 |
+
"grad_norm": 1.2890625,
|
| 1146 |
+
"learning_rate": 1.826625386996904e-05,
|
| 1147 |
+
"loss": 0.5474,
|
| 1148 |
+
"step": 1330
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 0.6575073601570167,
|
| 1152 |
+
"grad_norm": 1.4375,
|
| 1153 |
+
"learning_rate": 1.800825593395253e-05,
|
| 1154 |
+
"loss": 0.5195,
|
| 1155 |
+
"step": 1340
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 0.662414131501472,
|
| 1159 |
+
"grad_norm": 1.375,
|
| 1160 |
+
"learning_rate": 1.7750257997936018e-05,
|
| 1161 |
+
"loss": 0.4843,
|
| 1162 |
+
"step": 1350
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 0.662414131501472,
|
| 1166 |
+
"eval_loss": 0.5355111360549927,
|
| 1167 |
+
"eval_runtime": 28.5183,
|
| 1168 |
+
"eval_samples_per_second": 7.013,
|
| 1169 |
+
"eval_steps_per_second": 1.753,
|
| 1170 |
+
"step": 1350
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.6673209028459274,
|
| 1174 |
+
"grad_norm": 1.4140625,
|
| 1175 |
+
"learning_rate": 1.7492260061919503e-05,
|
| 1176 |
+
"loss": 0.5007,
|
| 1177 |
+
"step": 1360
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.6722276741903828,
|
| 1181 |
+
"grad_norm": 1.640625,
|
| 1182 |
+
"learning_rate": 1.7234262125902996e-05,
|
| 1183 |
+
"loss": 0.5328,
|
| 1184 |
+
"step": 1370
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.677134445534838,
|
| 1188 |
+
"grad_norm": 0.921875,
|
| 1189 |
+
"learning_rate": 1.697626418988648e-05,
|
| 1190 |
+
"loss": 0.5312,
|
| 1191 |
+
"step": 1380
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 0.6820412168792934,
|
| 1195 |
+
"grad_norm": 1.8359375,
|
| 1196 |
+
"learning_rate": 1.671826625386997e-05,
|
| 1197 |
+
"loss": 0.5863,
|
| 1198 |
+
"step": 1390
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.6869479882237488,
|
| 1202 |
+
"grad_norm": 0.87109375,
|
| 1203 |
+
"learning_rate": 1.646026831785346e-05,
|
| 1204 |
+
"loss": 0.5408,
|
| 1205 |
+
"step": 1400
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.6869479882237488,
|
| 1209 |
+
"eval_loss": 0.5326699614524841,
|
| 1210 |
+
"eval_runtime": 28.6461,
|
| 1211 |
+
"eval_samples_per_second": 6.982,
|
| 1212 |
+
"eval_steps_per_second": 1.745,
|
| 1213 |
+
"step": 1400
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6918547595682041,
|
| 1217 |
+
"grad_norm": 1.203125,
|
| 1218 |
+
"learning_rate": 1.6202270381836944e-05,
|
| 1219 |
+
"loss": 0.4963,
|
| 1220 |
+
"step": 1410
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6967615309126595,
|
| 1224 |
+
"grad_norm": 1.2734375,
|
| 1225 |
+
"learning_rate": 1.5944272445820436e-05,
|
| 1226 |
+
"loss": 0.5244,
|
| 1227 |
+
"step": 1420
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.7016683022571149,
|
| 1231 |
+
"grad_norm": 1.328125,
|
| 1232 |
+
"learning_rate": 1.568627450980392e-05,
|
| 1233 |
+
"loss": 0.5301,
|
| 1234 |
+
"step": 1430
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.7065750736015701,
|
| 1238 |
+
"grad_norm": 0.8984375,
|
| 1239 |
+
"learning_rate": 1.542827657378741e-05,
|
| 1240 |
+
"loss": 0.4815,
|
| 1241 |
+
"step": 1440
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.7114818449460255,
|
| 1245 |
+
"grad_norm": 0.56640625,
|
| 1246 |
+
"learning_rate": 1.5170278637770899e-05,
|
| 1247 |
+
"loss": 0.5156,
|
| 1248 |
+
"step": 1450
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.7114818449460255,
|
| 1252 |
+
"eval_loss": 0.532850980758667,
|
| 1253 |
+
"eval_runtime": 28.7098,
|
| 1254 |
+
"eval_samples_per_second": 6.966,
|
| 1255 |
+
"eval_steps_per_second": 1.742,
|
| 1256 |
+
"step": 1450
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7163886162904809,
|
| 1260 |
+
"grad_norm": 1.203125,
|
| 1261 |
+
"learning_rate": 1.4912280701754386e-05,
|
| 1262 |
+
"loss": 0.5187,
|
| 1263 |
+
"step": 1460
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.7212953876349362,
|
| 1267 |
+
"grad_norm": 2.828125,
|
| 1268 |
+
"learning_rate": 1.4654282765737876e-05,
|
| 1269 |
+
"loss": 0.4884,
|
| 1270 |
+
"step": 1470
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.7262021589793916,
|
| 1274 |
+
"grad_norm": 1.2109375,
|
| 1275 |
+
"learning_rate": 1.4396284829721363e-05,
|
| 1276 |
+
"loss": 0.5434,
|
| 1277 |
+
"step": 1480
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.7311089303238469,
|
| 1281 |
+
"grad_norm": 0.96875,
|
| 1282 |
+
"learning_rate": 1.4138286893704852e-05,
|
| 1283 |
+
"loss": 0.5154,
|
| 1284 |
+
"step": 1490
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7360157016683022,
|
| 1288 |
+
"grad_norm": 1.1640625,
|
| 1289 |
+
"learning_rate": 1.388028895768834e-05,
|
| 1290 |
+
"loss": 0.5312,
|
| 1291 |
+
"step": 1500
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7360157016683022,
|
| 1295 |
+
"eval_loss": 0.5351821184158325,
|
| 1296 |
+
"eval_runtime": 28.5695,
|
| 1297 |
+
"eval_samples_per_second": 7.0,
|
| 1298 |
+
"eval_steps_per_second": 1.75,
|
| 1299 |
+
"step": 1500
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.7409224730127576,
|
| 1303 |
+
"grad_norm": 1.328125,
|
| 1304 |
+
"learning_rate": 1.3622291021671826e-05,
|
| 1305 |
+
"loss": 0.5394,
|
| 1306 |
+
"step": 1510
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.745829244357213,
|
| 1310 |
+
"grad_norm": 2.078125,
|
| 1311 |
+
"learning_rate": 1.3364293085655317e-05,
|
| 1312 |
+
"loss": 0.5081,
|
| 1313 |
+
"step": 1520
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 0.7507360157016683,
|
| 1317 |
+
"grad_norm": 1.09375,
|
| 1318 |
+
"learning_rate": 1.3106295149638804e-05,
|
| 1319 |
+
"loss": 0.5498,
|
| 1320 |
+
"step": 1530
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 0.7556427870461236,
|
| 1324 |
+
"grad_norm": 0.71484375,
|
| 1325 |
+
"learning_rate": 1.2848297213622292e-05,
|
| 1326 |
+
"loss": 0.4704,
|
| 1327 |
+
"step": 1540
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 0.760549558390579,
|
| 1331 |
+
"grad_norm": 1.0078125,
|
| 1332 |
+
"learning_rate": 1.259029927760578e-05,
|
| 1333 |
+
"loss": 0.5095,
|
| 1334 |
+
"step": 1550
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 0.760549558390579,
|
| 1338 |
+
"eval_loss": 0.5329477787017822,
|
| 1339 |
+
"eval_runtime": 28.5741,
|
| 1340 |
+
"eval_samples_per_second": 6.999,
|
| 1341 |
+
"eval_steps_per_second": 1.75,
|
| 1342 |
+
"step": 1550
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 0.7654563297350343,
|
| 1346 |
+
"grad_norm": 0.86328125,
|
| 1347 |
+
"learning_rate": 1.2332301341589268e-05,
|
| 1348 |
+
"loss": 0.5021,
|
| 1349 |
+
"step": 1560
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 0.7703631010794897,
|
| 1353 |
+
"grad_norm": 1.3671875,
|
| 1354 |
+
"learning_rate": 1.2074303405572757e-05,
|
| 1355 |
+
"loss": 0.5578,
|
| 1356 |
+
"step": 1570
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 0.7752698724239451,
|
| 1360 |
+
"grad_norm": 1.296875,
|
| 1361 |
+
"learning_rate": 1.1816305469556244e-05,
|
| 1362 |
+
"loss": 0.5031,
|
| 1363 |
+
"step": 1580
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 0.7801766437684003,
|
| 1367 |
+
"grad_norm": 0.87890625,
|
| 1368 |
+
"learning_rate": 1.1558307533539733e-05,
|
| 1369 |
+
"loss": 0.5225,
|
| 1370 |
+
"step": 1590
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 0.7850834151128557,
|
| 1374 |
+
"grad_norm": 1.34375,
|
| 1375 |
+
"learning_rate": 1.130030959752322e-05,
|
| 1376 |
+
"loss": 0.4909,
|
| 1377 |
+
"step": 1600
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 0.7850834151128557,
|
| 1381 |
+
"eval_loss": 0.5333244204521179,
|
| 1382 |
+
"eval_runtime": 28.4593,
|
| 1383 |
+
"eval_samples_per_second": 7.028,
|
| 1384 |
+
"eval_steps_per_second": 1.757,
|
| 1385 |
+
"step": 1600
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 0.7899901864573111,
|
| 1389 |
+
"grad_norm": 1.2734375,
|
| 1390 |
+
"learning_rate": 1.1042311661506709e-05,
|
| 1391 |
+
"loss": 0.5154,
|
| 1392 |
+
"step": 1610
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 0.7948969578017664,
|
| 1396 |
+
"grad_norm": 2.0,
|
| 1397 |
+
"learning_rate": 1.0784313725490197e-05,
|
| 1398 |
+
"loss": 0.5048,
|
| 1399 |
+
"step": 1620
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 0.7998037291462218,
|
| 1403 |
+
"grad_norm": 1.078125,
|
| 1404 |
+
"learning_rate": 1.0526315789473684e-05,
|
| 1405 |
+
"loss": 0.4927,
|
| 1406 |
+
"step": 1630
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 0.8047105004906772,
|
| 1410 |
+
"grad_norm": 1.2265625,
|
| 1411 |
+
"learning_rate": 1.0268317853457173e-05,
|
| 1412 |
+
"loss": 0.5636,
|
| 1413 |
+
"step": 1640
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 0.8096172718351324,
|
| 1417 |
+
"grad_norm": 1.1796875,
|
| 1418 |
+
"learning_rate": 1.001031991744066e-05,
|
| 1419 |
+
"loss": 0.5423,
|
| 1420 |
+
"step": 1650
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 0.8096172718351324,
|
| 1424 |
+
"eval_loss": 0.5323337316513062,
|
| 1425 |
+
"eval_runtime": 28.6741,
|
| 1426 |
+
"eval_samples_per_second": 6.975,
|
| 1427 |
+
"eval_steps_per_second": 1.744,
|
| 1428 |
+
"step": 1650
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 0.8145240431795878,
|
| 1432 |
+
"grad_norm": 1.546875,
|
| 1433 |
+
"learning_rate": 9.752321981424149e-06,
|
| 1434 |
+
"loss": 0.4747,
|
| 1435 |
+
"step": 1660
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 0.8194308145240432,
|
| 1439 |
+
"grad_norm": 1.0078125,
|
| 1440 |
+
"learning_rate": 9.494324045407638e-06,
|
| 1441 |
+
"loss": 0.4759,
|
| 1442 |
+
"step": 1670
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.8243375858684985,
|
| 1446 |
+
"grad_norm": 0.87890625,
|
| 1447 |
+
"learning_rate": 9.236326109391125e-06,
|
| 1448 |
+
"loss": 0.5072,
|
| 1449 |
+
"step": 1680
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.8292443572129539,
|
| 1453 |
+
"grad_norm": 0.83984375,
|
| 1454 |
+
"learning_rate": 8.978328173374614e-06,
|
| 1455 |
+
"loss": 0.4662,
|
| 1456 |
+
"step": 1690
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 0.8341511285574092,
|
| 1460 |
+
"grad_norm": 0.95703125,
|
| 1461 |
+
"learning_rate": 8.7203302373581e-06,
|
| 1462 |
+
"loss": 0.5161,
|
| 1463 |
+
"step": 1700
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 0.8341511285574092,
|
| 1467 |
+
"eval_loss": 0.5313096046447754,
|
| 1468 |
+
"eval_runtime": 28.6753,
|
| 1469 |
+
"eval_samples_per_second": 6.975,
|
| 1470 |
+
"eval_steps_per_second": 1.744,
|
| 1471 |
+
"step": 1700
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 0.8390578999018645,
|
| 1475 |
+
"grad_norm": 1.3203125,
|
| 1476 |
+
"learning_rate": 8.46233230134159e-06,
|
| 1477 |
+
"loss": 0.4837,
|
| 1478 |
+
"step": 1710
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 0.8439646712463199,
|
| 1482 |
+
"grad_norm": 1.1171875,
|
| 1483 |
+
"learning_rate": 8.204334365325078e-06,
|
| 1484 |
+
"loss": 0.49,
|
| 1485 |
+
"step": 1720
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 0.8488714425907753,
|
| 1489 |
+
"grad_norm": 1.515625,
|
| 1490 |
+
"learning_rate": 7.946336429308567e-06,
|
| 1491 |
+
"loss": 0.5959,
|
| 1492 |
+
"step": 1730
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 0.8537782139352306,
|
| 1496 |
+
"grad_norm": 1.1171875,
|
| 1497 |
+
"learning_rate": 7.688338493292054e-06,
|
| 1498 |
+
"loss": 0.5113,
|
| 1499 |
+
"step": 1740
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 0.858684985279686,
|
| 1503 |
+
"grad_norm": 1.4375,
|
| 1504 |
+
"learning_rate": 7.430340557275542e-06,
|
| 1505 |
+
"loss": 0.5343,
|
| 1506 |
+
"step": 1750
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 0.858684985279686,
|
| 1510 |
+
"eval_loss": 0.530957043170929,
|
| 1511 |
+
"eval_runtime": 28.7519,
|
| 1512 |
+
"eval_samples_per_second": 6.956,
|
| 1513 |
+
"eval_steps_per_second": 1.739,
|
| 1514 |
+
"step": 1750
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.8635917566241413,
|
| 1518 |
+
"grad_norm": 1.109375,
|
| 1519 |
+
"learning_rate": 7.1723426212590306e-06,
|
| 1520 |
+
"loss": 0.5282,
|
| 1521 |
+
"step": 1760
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.8684985279685966,
|
| 1525 |
+
"grad_norm": 1.5,
|
| 1526 |
+
"learning_rate": 6.9143446852425185e-06,
|
| 1527 |
+
"loss": 0.4993,
|
| 1528 |
+
"step": 1770
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.873405299313052,
|
| 1532 |
+
"grad_norm": 1.1328125,
|
| 1533 |
+
"learning_rate": 6.656346749226007e-06,
|
| 1534 |
+
"loss": 0.4911,
|
| 1535 |
+
"step": 1780
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.8783120706575074,
|
| 1539 |
+
"grad_norm": 1.1640625,
|
| 1540 |
+
"learning_rate": 6.398348813209494e-06,
|
| 1541 |
+
"loss": 0.4945,
|
| 1542 |
+
"step": 1790
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.8832188420019627,
|
| 1546 |
+
"grad_norm": 0.6796875,
|
| 1547 |
+
"learning_rate": 6.140350877192982e-06,
|
| 1548 |
+
"loss": 0.5435,
|
| 1549 |
+
"step": 1800
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.8832188420019627,
|
| 1553 |
+
"eval_loss": 0.5307140350341797,
|
| 1554 |
+
"eval_runtime": 28.6268,
|
| 1555 |
+
"eval_samples_per_second": 6.986,
|
| 1556 |
+
"eval_steps_per_second": 1.747,
|
| 1557 |
+
"step": 1800
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.888125613346418,
|
| 1561 |
+
"grad_norm": 0.7890625,
|
| 1562 |
+
"learning_rate": 5.882352941176471e-06,
|
| 1563 |
+
"loss": 0.5001,
|
| 1564 |
+
"step": 1810
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8930323846908734,
|
| 1568 |
+
"grad_norm": 1.265625,
|
| 1569 |
+
"learning_rate": 5.624355005159959e-06,
|
| 1570 |
+
"loss": 0.5042,
|
| 1571 |
+
"step": 1820
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8979391560353287,
|
| 1575 |
+
"grad_norm": 1.1875,
|
| 1576 |
+
"learning_rate": 5.366357069143447e-06,
|
| 1577 |
+
"loss": 0.5642,
|
| 1578 |
+
"step": 1830
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.9028459273797841,
|
| 1582 |
+
"grad_norm": 1.2265625,
|
| 1583 |
+
"learning_rate": 5.1083591331269355e-06,
|
| 1584 |
+
"loss": 0.5587,
|
| 1585 |
+
"step": 1840
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.9077526987242395,
|
| 1589 |
+
"grad_norm": 1.921875,
|
| 1590 |
+
"learning_rate": 4.850361197110423e-06,
|
| 1591 |
+
"loss": 0.539,
|
| 1592 |
+
"step": 1850
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.9077526987242395,
|
| 1596 |
+
"eval_loss": 0.5304572582244873,
|
| 1597 |
+
"eval_runtime": 28.4933,
|
| 1598 |
+
"eval_samples_per_second": 7.019,
|
| 1599 |
+
"eval_steps_per_second": 1.755,
|
| 1600 |
+
"step": 1850
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.9126594700686947,
|
| 1604 |
+
"grad_norm": 1.375,
|
| 1605 |
+
"learning_rate": 4.592363261093911e-06,
|
| 1606 |
+
"loss": 0.5004,
|
| 1607 |
+
"step": 1860
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 0.9175662414131501,
|
| 1611 |
+
"grad_norm": 1.234375,
|
| 1612 |
+
"learning_rate": 4.3343653250774e-06,
|
| 1613 |
+
"loss": 0.4948,
|
| 1614 |
+
"step": 1870
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.9224730127576055,
|
| 1618 |
+
"grad_norm": 1.2109375,
|
| 1619 |
+
"learning_rate": 4.076367389060888e-06,
|
| 1620 |
+
"loss": 0.5392,
|
| 1621 |
+
"step": 1880
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 0.9273797841020608,
|
| 1625 |
+
"grad_norm": 1.1171875,
|
| 1626 |
+
"learning_rate": 3.818369453044376e-06,
|
| 1627 |
+
"loss": 0.5441,
|
| 1628 |
+
"step": 1890
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 0.9322865554465162,
|
| 1632 |
+
"grad_norm": 2.65625,
|
| 1633 |
+
"learning_rate": 3.560371517027864e-06,
|
| 1634 |
+
"loss": 0.5096,
|
| 1635 |
+
"step": 1900
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 0.9322865554465162,
|
| 1639 |
+
"eval_loss": 0.5307794809341431,
|
| 1640 |
+
"eval_runtime": 28.4936,
|
| 1641 |
+
"eval_samples_per_second": 7.019,
|
| 1642 |
+
"eval_steps_per_second": 1.755,
|
| 1643 |
+
"step": 1900
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 0.9371933267909716,
|
| 1647 |
+
"grad_norm": 0.46875,
|
| 1648 |
+
"learning_rate": 3.3023735810113516e-06,
|
| 1649 |
+
"loss": 0.5092,
|
| 1650 |
+
"step": 1910
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 0.9421000981354269,
|
| 1654 |
+
"grad_norm": 1.0546875,
|
| 1655 |
+
"learning_rate": 3.0443756449948404e-06,
|
| 1656 |
+
"loss": 0.5825,
|
| 1657 |
+
"step": 1920
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 0.9470068694798822,
|
| 1661 |
+
"grad_norm": 1.5078125,
|
| 1662 |
+
"learning_rate": 2.7863777089783283e-06,
|
| 1663 |
+
"loss": 0.5331,
|
| 1664 |
+
"step": 1930
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 0.9519136408243376,
|
| 1668 |
+
"grad_norm": 1.03125,
|
| 1669 |
+
"learning_rate": 2.5283797729618166e-06,
|
| 1670 |
+
"loss": 0.5464,
|
| 1671 |
+
"step": 1940
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 0.956820412168793,
|
| 1675 |
+
"grad_norm": 1.1640625,
|
| 1676 |
+
"learning_rate": 2.2703818369453045e-06,
|
| 1677 |
+
"loss": 0.5155,
|
| 1678 |
+
"step": 1950
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 0.956820412168793,
|
| 1682 |
+
"eval_loss": 0.5306495428085327,
|
| 1683 |
+
"eval_runtime": 28.7357,
|
| 1684 |
+
"eval_samples_per_second": 6.96,
|
| 1685 |
+
"eval_steps_per_second": 1.74,
|
| 1686 |
+
"step": 1950
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 0.9617271835132483,
|
| 1690 |
+
"grad_norm": 0.9921875,
|
| 1691 |
+
"learning_rate": 2.012383900928793e-06,
|
| 1692 |
+
"loss": 0.5282,
|
| 1693 |
+
"step": 1960
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 0.9666339548577036,
|
| 1697 |
+
"grad_norm": 0.40234375,
|
| 1698 |
+
"learning_rate": 1.7543859649122807e-06,
|
| 1699 |
+
"loss": 0.5208,
|
| 1700 |
+
"step": 1970
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 0.971540726202159,
|
| 1704 |
+
"grad_norm": 0.78515625,
|
| 1705 |
+
"learning_rate": 1.4963880288957689e-06,
|
| 1706 |
+
"loss": 0.5216,
|
| 1707 |
+
"step": 1980
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 0.9764474975466143,
|
| 1711 |
+
"grad_norm": 1.3359375,
|
| 1712 |
+
"learning_rate": 1.2383900928792572e-06,
|
| 1713 |
+
"loss": 0.516,
|
| 1714 |
+
"step": 1990
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 0.9813542688910697,
|
| 1718 |
+
"grad_norm": 3.125,
|
| 1719 |
+
"learning_rate": 9.80392156862745e-07,
|
| 1720 |
+
"loss": 0.529,
|
| 1721 |
+
"step": 2000
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 0.9813542688910697,
|
| 1725 |
+
"eval_loss": 0.5309420228004456,
|
| 1726 |
+
"eval_runtime": 28.7264,
|
| 1727 |
+
"eval_samples_per_second": 6.962,
|
| 1728 |
+
"eval_steps_per_second": 1.741,
|
| 1729 |
+
"step": 2000
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 0.9862610402355251,
|
| 1733 |
+
"grad_norm": 1.3046875,
|
| 1734 |
+
"learning_rate": 7.223942208462333e-07,
|
| 1735 |
+
"loss": 0.5101,
|
| 1736 |
+
"step": 2010
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 0.9911678115799804,
|
| 1740 |
+
"grad_norm": 1.109375,
|
| 1741 |
+
"learning_rate": 4.6439628482972136e-07,
|
| 1742 |
+
"loss": 0.4883,
|
| 1743 |
+
"step": 2020
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 0.9960745829244357,
|
| 1747 |
+
"grad_norm": 0.71484375,
|
| 1748 |
+
"learning_rate": 2.0639834881320948e-07,
|
| 1749 |
+
"loss": 0.5151,
|
| 1750 |
+
"step": 2030
|
| 1751 |
+
}
|
| 1752 |
+
],
|
| 1753 |
+
"logging_steps": 10,
|
| 1754 |
+
"max_steps": 2038,
|
| 1755 |
+
"num_input_tokens_seen": 0,
|
| 1756 |
+
"num_train_epochs": 1,
|
| 1757 |
+
"save_steps": 500,
|
| 1758 |
+
"stateful_callbacks": {
|
| 1759 |
+
"TrainerControl": {
|
| 1760 |
+
"args": {
|
| 1761 |
+
"should_epoch_stop": false,
|
| 1762 |
+
"should_evaluate": false,
|
| 1763 |
+
"should_log": false,
|
| 1764 |
+
"should_save": true,
|
| 1765 |
+
"should_training_stop": true
|
| 1766 |
+
},
|
| 1767 |
+
"attributes": {}
|
| 1768 |
+
}
|
| 1769 |
+
},
|
| 1770 |
+
"total_flos": 3.79578628062624e+16,
|
| 1771 |
+
"train_batch_size": 4,
|
| 1772 |
+
"trial_name": null,
|
| 1773 |
+
"trial_params": null
|
| 1774 |
+
}
|
checkpoint-2038/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34c4857badce80c70f064f21ad667b2f5f1313b4d2fe2b5a59ca22c17dde065c
|
| 3 |
+
size 5176
|
git_hash.txt
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
55b2dde274ae08820ed112e2e52dcff41853bd34
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_seq_length": 256,
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"processor_class": "PaliGemmaProcessor",
|
| 19 |
+
"resample": 3,
|
| 20 |
+
"rescale_factor": 0.00392156862745098,
|
| 21 |
+
"size": {
|
| 22 |
+
"height": 224,
|
| 23 |
+
"width": 224
|
| 24 |
+
}
|
| 25 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
{
|
| 4 |
+
"content": "<image>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
}
|
| 10 |
+
],
|
| 11 |
+
"bos_token": {
|
| 12 |
+
"content": "<bos>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false
|
| 17 |
+
},
|
| 18 |
+
"eos_token": {
|
| 19 |
+
"content": "<eos>",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
},
|
| 25 |
+
"pad_token": {
|
| 26 |
+
"content": "<pad>",
|
| 27 |
+
"lstrip": false,
|
| 28 |
+
"normalized": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"single_word": false
|
| 31 |
+
},
|
| 32 |
+
"unk_token": {
|
| 33 |
+
"content": "<unk>",
|
| 34 |
+
"lstrip": false,
|
| 35 |
+
"normalized": false,
|
| 36 |
+
"rstrip": false,
|
| 37 |
+
"single_word": false
|
| 38 |
+
}
|
| 39 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1df2ab04780faccf51a881d7c5a7026cc6f979083af2eebf709d051b8d47134b
|
| 3 |
+
size 17763458
|
tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
training_config.yml
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
config:
|
| 2 |
+
(): colpali_engine.utils.train_colpali_engine_models.ColModelTrainingConfig
|
| 3 |
+
output_dir: !path ../../../models/without_tabfquad/train_colpali-3b-ft-ocrvqa-224
|
| 4 |
+
processor:
|
| 5 |
+
() : colpali_engine.utils.wrapper.AutoProcessorWrapper
|
| 6 |
+
pretrained_model_name_or_path: "google/paligemma-3b-ft-ocrvqa-224"
|
| 7 |
+
max_length: 50
|
| 8 |
+
model:
|
| 9 |
+
(): colpali_engine.utils.wrapper.AutoColModelWrapper
|
| 10 |
+
pretrained_model_name_or_path: "google/paligemma-3b-ft-ocrvqa-224"
|
| 11 |
+
training_objective: "colbertv1"
|
| 12 |
+
# attn_implementation: "eager"
|
| 13 |
+
# attn_implementation: "flash_attention_2"
|
| 14 |
+
torch_dtype: !ext torch.bfloat16
|
| 15 |
+
# device_map: "auto"
|
| 16 |
+
# quantization_config:
|
| 17 |
+
# (): transformers.BitsAndBytesConfig
|
| 18 |
+
# load_in_4bit: true
|
| 19 |
+
# bnb_4bit_quant_type: "nf4"
|
| 20 |
+
# bnb_4bit_compute_dtype: "bfloat16"
|
| 21 |
+
# bnb_4bit_use_double_quant: true
|
| 22 |
+
|
| 23 |
+
dataset_loading_func: !ext colpali_engine.utils.dataset_transformation.load_docvqa_dataset
|
| 24 |
+
eval_dataset_loader: !import ../data/test_data.yaml
|
| 25 |
+
|
| 26 |
+
max_length: 50
|
| 27 |
+
run_eval: true
|
| 28 |
+
add_suffix: true
|
| 29 |
+
loss_func:
|
| 30 |
+
(): colpali_engine.loss.colbert_loss.XtrPairwiseCELoss
|
| 31 |
+
tr_args: !import ../tr_args/default_tr_args.yaml
|
| 32 |
+
peft_config:
|
| 33 |
+
(): peft.LoraConfig
|
| 34 |
+
r: 32
|
| 35 |
+
lora_alpha: 32
|
| 36 |
+
lora_dropout: 0.1
|
| 37 |
+
init_lora_weights: "gaussian"
|
| 38 |
+
bias: "none"
|
| 39 |
+
task_type: "FEATURE_EXTRACTION"
|
| 40 |
+
target_modules: '(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
|
| 41 |
+
# target_modules: '(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
|
| 42 |
+
|