Dar4devil commited on
Commit
7f3188b
·
1 Parent(s): 08b60ff

Add completed GRPO training evidence and pushed checkpoint

Browse files
outputs/logs/hf_job_pushed_inspect.txt ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "id": "69ed188dd2c8bd8662bce4e7",
4
+ "created_at": "2026-04-25 19:39:57.695000+00:00",
5
+ "docker_image": "ghcr.io/astral-sh/uv:python3.12-bookworm",
6
+ "space_id": null,
7
+ "command": [
8
+ "uv",
9
+ "run",
10
+ "--with",
11
+ "incident-commander-env[train] @ git+https://huggingface.co/spaces/Dar3devil/incident_commander_env",
12
+ "--with",
13
+ "transformers @ git+https://github.com/huggingface/transformers.git@main",
14
+ "--with",
15
+ "trl @ git+https://github.com/huggingface/trl.git@main",
16
+ "--with",
17
+ "jmespath",
18
+ "incident-commander-train",
19
+ "--model",
20
+ "Qwen/Qwen3-0.6B",
21
+ "--output-dir",
22
+ "outputs/grpo-qwen3-0.6b-pushed",
23
+ "--max-steps",
24
+ "120",
25
+ "--num-generations",
26
+ "2",
27
+ "--gradient-accumulation-steps",
28
+ "8",
29
+ "--no-vllm",
30
+ "--push-to-hub",
31
+ "--hub-model-id",
32
+ "Dar3devil/incident-commander-qwen3-0.6b-grpo"
33
+ ],
34
+ "arguments": [],
35
+ "environment": {},
36
+ "secrets": [
37
+ "HF_TOKEN"
38
+ ],
39
+ "flavor": "a10g-large",
40
+ "labels": {},
41
+ "volumes": null,
42
+ "status": {
43
+ "stage": "COMPLETED",
44
+ "message": null
45
+ },
46
+ "owner": {
47
+ "id": "69d7942af64e853dd849884a",
48
+ "name": "Dar3devil",
49
+ "type": "user"
50
+ },
51
+ "endpoint": "https://huggingface.co",
52
+ "url": "https://huggingface.co/jobs/Dar3devil/69ed188dd2c8bd8662bce4e7"
53
+ }
54
+ ]