Upload full trainer checkpoint step=1173 (profile=llama3_2_1b_base_grpo, run_type=grpo)
Browse files- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/adapter_config.json +2 -2
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/optimizer.pt +1 -1
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/resume_sft_best/adapter_config.json +5 -5
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/resume_sft_best/adapter_model.safetensors +1 -1
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/trainer_state.json +0 -0
- meta-llama__llama-3.2-1b/grpo/run_manifest.json +11 -17
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/adapter_config.json
CHANGED
|
@@ -33,12 +33,12 @@
|
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
|
|
|
|
|
|
| 36 |
"up_proj",
|
| 37 |
"q_proj",
|
| 38 |
"down_proj",
|
| 39 |
-
"v_proj",
|
| 40 |
"k_proj",
|
| 41 |
-
"o_proj",
|
| 42 |
"gate_proj"
|
| 43 |
],
|
| 44 |
"target_parameters": null,
|
|
|
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
| 36 |
+
"o_proj",
|
| 37 |
+
"v_proj",
|
| 38 |
"up_proj",
|
| 39 |
"q_proj",
|
| 40 |
"down_proj",
|
|
|
|
| 41 |
"k_proj",
|
|
|
|
| 42 |
"gate_proj"
|
| 43 |
],
|
| 44 |
"target_parameters": null,
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 23162187
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a092262c13f71bf52afaf6a968737c0ec0f589894f0eb927e55be5b435ace6c
|
| 3 |
size 23162187
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/resume_sft_best/adapter_config.json
CHANGED
|
@@ -33,13 +33,13 @@
|
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
| 36 |
-
"
|
| 37 |
-
"
|
| 38 |
"down_proj",
|
| 39 |
-
"
|
|
|
|
| 40 |
"k_proj",
|
| 41 |
-
"
|
| 42 |
-
"gate_proj"
|
| 43 |
],
|
| 44 |
"target_parameters": null,
|
| 45 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
| 36 |
+
"o_proj",
|
| 37 |
+
"gate_proj",
|
| 38 |
"down_proj",
|
| 39 |
+
"q_proj",
|
| 40 |
+
"up_proj",
|
| 41 |
"k_proj",
|
| 42 |
+
"v_proj"
|
|
|
|
| 43 |
],
|
| 44 |
"target_parameters": null,
|
| 45 |
"task_type": "CAUSAL_LM",
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/resume_sft_best/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 45118424
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0fc0173538c065e3a85d333561f7bce9aac6efb38d274821cab522adbcec03b
|
| 3 |
size 45118424
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173/trainer_state.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
meta-llama__llama-3.2-1b/grpo/run_manifest.json
CHANGED
|
@@ -1,31 +1,25 @@
|
|
| 1 |
{
|
| 2 |
"schema_version": 1,
|
| 3 |
-
"run_id": "
|
| 4 |
"run_type": "grpo",
|
| 5 |
"profile_name": "llama3_2_1b_base_grpo",
|
| 6 |
"model_name": "meta-llama/Llama-3.2-1B",
|
| 7 |
"status": "running",
|
| 8 |
"best_metric_name": "eval_cer_subset",
|
| 9 |
-
"best_metric_value": 0.
|
| 10 |
-
"best_step":
|
| 11 |
-
"best_checkpoint_step":
|
| 12 |
-
"latest_checkpoint_step":
|
| 13 |
"checkpoints": [
|
| 14 |
{
|
| 15 |
-
"step":
|
| 16 |
-
"epoch":
|
| 17 |
-
"repo_path": "meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-
|
| 18 |
-
"timestamp": "2026-04-
|
| 19 |
-
},
|
| 20 |
-
{
|
| 21 |
-
"step": 18,
|
| 22 |
-
"epoch": 9.591610537982778e-05,
|
| 23 |
-
"repo_path": "meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-18",
|
| 24 |
-
"timestamp": "2026-04-08T20:49:25.651833+00:00"
|
| 25 |
}
|
| 26 |
],
|
| 27 |
"epoch_artifacts": [],
|
| 28 |
"resume_history": [],
|
| 29 |
-
"created_at": "2026-04-08T20:
|
| 30 |
-
"updated_at": "2026-04-
|
| 31 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"schema_version": 1,
|
| 3 |
+
"run_id": "44a9bcf9-66a6-4b97-a131-62881f53fa88",
|
| 4 |
"run_type": "grpo",
|
| 5 |
"profile_name": "llama3_2_1b_base_grpo",
|
| 6 |
"model_name": "meta-llama/Llama-3.2-1B",
|
| 7 |
"status": "running",
|
| 8 |
"best_metric_name": "eval_cer_subset",
|
| 9 |
+
"best_metric_value": 0.012881267607927434,
|
| 10 |
+
"best_step": 1173,
|
| 11 |
+
"best_checkpoint_step": 1173,
|
| 12 |
+
"latest_checkpoint_step": 1173,
|
| 13 |
"checkpoints": [
|
| 14 |
{
|
| 15 |
+
"step": 1173,
|
| 16 |
+
"epoch": 0.00625053286725211,
|
| 17 |
+
"repo_path": "meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-1173",
|
| 18 |
+
"timestamp": "2026-04-08T21:07:07.041384+00:00"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
}
|
| 20 |
],
|
| 21 |
"epoch_artifacts": [],
|
| 22 |
"resume_history": [],
|
| 23 |
+
"created_at": "2026-04-08T20:52:57.531787+00:00",
|
| 24 |
+
"updated_at": "2026-04-08T21:07:07.041401+00:00"
|
| 25 |
}
|