Hugging Face's logo

JonesLin
/
next-jev-stage2-last-loop

next-jev
workspace-reasoning
nli
cot
Model card Files
xet
 Community
next-jev-stage2-last-loop
File size: 2,522 Bytes
2f0aa38
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
{
  "run_id": "f4acad907e444bf79d25f05e1d61aab2",
  "model": {
    "model_name_or_path": "models/Qwen3.5-0.8B",
    "revision": "2fc06364715b967f1860aea9cf38778875588b17",
    "dtype": "bfloat16",
    "attn_implementation": "sdpa",
    "workspace_tokens": 128,
    "loops": 3,
    "loop_scope": "middle_workspace",
    "loop_start_block": 1,
    "loop_end_block": 24,
    "detach_loop_state": false,
    "classification_loss_scope": "last_loop",
    "aux_memory_mode": "last_loop",
    "aux_weight": 0.3,
    "aux_dim": 256,
    "aux_heads": 8,
    "aux_max_tokens": 4096,
    "aux_chunk_size": 1024,
    "aux_response_chunk_size": 16,
    "aux_layers": 1,
    "aux_self_attn_layers": null,
    "aux_self_attn_window": null,
    "freeze_vision": false,
    "gradient_checkpointing": true,
    "training_stage": 2,
    "stage1_objective": "answer_and_cot"
  },
  "train": {
    "train_file": "data/phase123-multi-response/train.jsonl",
    "validation_file": "data/phase123-multi-response/validation.jsonl",
    "output_dir": "runs/stage2-last-loop",
    "epochs": 1,
    "batch_size": 32,
    "gradient_accumulation": 1,
    "learning_rate": 8e-06,
    "lr_schedule": "warmup_cosine",
    "warmup_ratio": 0.0,
    "min_learning_rate": 8e-07,
    "weight_decay": 0.01,
    "max_grad_norm": 1.0,
    "max_length": 2048,
    "max_image_pixels": 262144,
    "seed": 42,
    "device": "cuda",
    "max_steps": null,
    "save_every": 100,
    "keep_checkpoints": 2,
    "num_workers": 4,
    "classifier_learning_rate": 8e-05,
    "init_checkpoint": "runs/stage1-multi-response/checkpoint-00006347",
    "answer_max_tokens": 256
  },
  "source_hashes": {
    "train_file": "52e712f8bf847f952515a4b377332c6765b97bcf4dfa8a82a06336fb56844408",
    "validation_file": "18e90d614a6da718fc9acec5571adcfb2c0003afbbf55d93b75d3fd47ad0e927",
    "train_images": "842bba11f1084b9c31cf9fc069c6e8310fd89e106e36c0c627e18a8c96ca8a1f",
    "validation_images": "b3a1754a3160fd19c88082a8294381c44a58058dfcdaea104e644eec0ccf30e8"
  },
  "learning_rate_plan": {
    "type": "warmup_cosine",
    "total_steps": 5921,
    "warmup_steps": 0,
    "initial_learning_rate": 8e-06,
    "min_learning_rate": 8e-07
  },
  "torch_version": "2.14.0+cu130",
  "transformers_version": "5.17.0",
  "parameter_dtype": "float32",
  "initialization": {
    "checkpoint": "/scratch/255028/next_jev/runs/stage1-multi-response/checkpoint-00006347",
    "sha256": "5fda55205f1855792ea60a8dfb5368766e6c861860e08f7e55609e18270c76f8",
    "reset_classifier": true
  }
}