Hugging Face's logo

PedramR
/
ASR_Post-processing

TensorBoard
Safetensors
Model card Files
xet
 Metrics Community
PedramR commited on
Commit
8331ebe
·
verified ·
1 Parent(s): 529fe08

checkpoint-100

Browse files
Files changed (7) hide show
  1. qwen3.5-2b/checkpoint-100/adapter_config.json +6 -6
  2. qwen3.5-2b/checkpoint-100/adapter_model.safetensors +1 -1
  3. qwen3.5-2b/checkpoint-100/optimizer.pt +1 -1
  4. qwen3.5-2b/checkpoint-100/trainer_state.json +24 -24
  5. qwen3.5-2b/tb/events.out.tfevents.1789940092.fed349279e2c.7945.0 +3 -0
  6. qwen3.5-2b/test_eval/metrics.json +1 -1
  7. qwen3.5-2b/test_eval/predictions.jsonl +0 -0
qwen3.5-2b/checkpoint-100/adapter_config.json CHANGED
@@ -32,18 +32,18 @@
32
  "rank_pattern": {},
33
  "revision": null,
34
  "target_modules": [
35
- "o_proj",
36
  "gate_proj",
37
- "in_proj_qkv",
38
  "in_proj_b",
39
  "out_proj",
40
- "q_proj",
41
  "k_proj",
 
 
42
  "in_proj_z",
43
- "up_proj",
44
  "v_proj",
45
- "in_proj_a",
46
- "down_proj"
 
 
47
  ],
48
  "target_parameters": null,
49
  "task_type": "CAUSAL_LM",
 
32
  "rank_pattern": {},
33
  "revision": null,
34
  "target_modules": [
 
35
  "gate_proj",
 
36
  "in_proj_b",
37
  "out_proj",
 
38
  "k_proj",
39
+ "o_proj",
40
+ "q_proj",
41
  "in_proj_z",
 
42
  "v_proj",
43
+ "up_proj",
44
+ "in_proj_qkv",
45
+ "down_proj",
46
+ "in_proj_a"
47
  ],
48
  "target_parameters": null,
49
  "task_type": "CAUSAL_LM",
qwen3.5-2b/checkpoint-100/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3516fb226a2b4bc6e564bc77e0f9bcf99ed02be7d1eb7622e36585ca32f548fe
3
  size 134604152
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b354598c9fedef3d5218417ffbf82c8ae665cf3c58e8ddbe8d685b8ddd06bf4e
3
  size 134604152
qwen3.5-2b/checkpoint-100/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fab5989da71e8335e77257636d2ba05a5e06e6c28915b4bcca735c064473e828
3
  size 269426431
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c25f3a565b4cb2dc34417c2eb57451e6c275e00d4a84d70916a3700bc38afaa
3
  size 269426431
qwen3.5-2b/checkpoint-100/trainer_state.json CHANGED
@@ -11,80 +11,80 @@
11
  "log_history": [
12
  {
13
  "epoch": 0.007914523149980214,
14
- "grad_norm": 1.111193060874939,
15
  "learning_rate": 1.4173228346456694e-05,
16
- "loss": 0.9034583091735839,
17
  "step": 10
18
  },
19
  {
20
  "epoch": 0.015829046299960427,
21
- "grad_norm": 0.8373973369598389,
22
  "learning_rate": 2.992125984251969e-05,
23
- "loss": 0.7779301166534424,
24
  "step": 20
25
  },
26
  {
27
  "epoch": 0.02374356944994064,
28
- "grad_norm": 0.7983554601669312,
29
  "learning_rate": 4.566929133858268e-05,
30
- "loss": 0.6945425987243652,
31
  "step": 30
32
  },
33
  {
34
  "epoch": 0.031658092599920855,
35
- "grad_norm": 0.7307385206222534,
36
  "learning_rate": 6.141732283464568e-05,
37
- "loss": 0.6987877368927002,
38
  "step": 40
39
  },
40
  {
41
  "epoch": 0.03957261574990107,
42
- "grad_norm": 1.0141053199768066,
43
  "learning_rate": 7.716535433070867e-05,
44
- "loss": 0.697762393951416,
45
  "step": 50
46
  },
47
  {
48
  "epoch": 0.04748713889988128,
49
- "grad_norm": 0.8023157715797424,
50
  "learning_rate": 9.291338582677166e-05,
51
- "loss": 0.5860283851623536,
52
  "step": 60
53
  },
54
  {
55
  "epoch": 0.055401662049861494,
56
- "grad_norm": 0.8767536282539368,
57
  "learning_rate": 0.00010866141732283466,
58
- "loss": 0.6935072422027588,
59
  "step": 70
60
  },
61
  {
62
  "epoch": 0.06331618519984171,
63
- "grad_norm": 0.7834345698356628,
64
  "learning_rate": 0.00012440944881889765,
65
- "loss": 0.6521744728088379,
66
  "step": 80
67
  },
68
  {
69
  "epoch": 0.07123070834982193,
70
- "grad_norm": 0.8145639896392822,
71
  "learning_rate": 0.00014015748031496062,
72
- "loss": 0.6522610664367676,
73
  "step": 90
74
  },
75
  {
76
  "epoch": 0.07914523149980214,
77
- "grad_norm": 0.9174225926399231,
78
  "learning_rate": 0.00015590551181102362,
79
- "loss": 0.6533907413482666,
80
  "step": 100
81
  },
82
  {
83
  "epoch": 0.07914523149980214,
84
- "eval_loss": 0.6276857256889343,
85
- "eval_runtime": 460.4363,
86
- "eval_samples_per_second": 24.327,
87
- "eval_steps_per_second": 6.083,
88
  "step": 100
89
  }
90
  ],
 
11
  "log_history": [
12
  {
13
  "epoch": 0.007914523149980214,
14
+ "grad_norm": 1.110097050666809,
15
  "learning_rate": 1.4173228346456694e-05,
16
+ "loss": 0.9039163589477539,
17
  "step": 10
18
  },
19
  {
20
  "epoch": 0.015829046299960427,
21
+ "grad_norm": 0.8214025497436523,
22
  "learning_rate": 2.992125984251969e-05,
23
+ "loss": 0.7784994125366211,
24
  "step": 20
25
  },
26
  {
27
  "epoch": 0.02374356944994064,
28
+ "grad_norm": 0.796338677406311,
29
  "learning_rate": 4.566929133858268e-05,
30
+ "loss": 0.694743537902832,
31
  "step": 30
32
  },
33
  {
34
  "epoch": 0.031658092599920855,
35
+ "grad_norm": 0.7282607555389404,
36
  "learning_rate": 6.141732283464568e-05,
37
+ "loss": 0.6986559391021728,
38
  "step": 40
39
  },
40
  {
41
  "epoch": 0.03957261574990107,
42
+ "grad_norm": 1.00419020652771,
43
  "learning_rate": 7.716535433070867e-05,
44
+ "loss": 0.697541332244873,
45
  "step": 50
46
  },
47
  {
48
  "epoch": 0.04748713889988128,
49
+ "grad_norm": 0.7905988097190857,
50
  "learning_rate": 9.291338582677166e-05,
51
+ "loss": 0.586260461807251,
52
  "step": 60
53
  },
54
  {
55
  "epoch": 0.055401662049861494,
56
+ "grad_norm": 0.8630598187446594,
57
  "learning_rate": 0.00010866141732283466,
58
+ "loss": 0.6937695503234863,
59
  "step": 70
60
  },
61
  {
62
  "epoch": 0.06331618519984171,
63
+ "grad_norm": 0.7757033705711365,
64
  "learning_rate": 0.00012440944881889765,
65
+ "loss": 0.652254295349121,
66
  "step": 80
67
  },
68
  {
69
  "epoch": 0.07123070834982193,
70
+ "grad_norm": 0.8161948919296265,
71
  "learning_rate": 0.00014015748031496062,
72
+ "loss": 0.6520013332366943,
73
  "step": 90
74
  },
75
  {
76
  "epoch": 0.07914523149980214,
77
+ "grad_norm": 0.8955326676368713,
78
  "learning_rate": 0.00015590551181102362,
79
+ "loss": 0.6528121948242187,
80
  "step": 100
81
  },
82
  {
83
  "epoch": 0.07914523149980214,
84
+ "eval_loss": 0.6273276209831238,
85
+ "eval_runtime": 460.2352,
86
+ "eval_samples_per_second": 24.338,
87
+ "eval_steps_per_second": 6.086,
88
  "step": 100
89
  }
90
  ],
qwen3.5-2b/tb/events.out.tfevents.1789940092.fed349279e2c.7945.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5edf34af6e6409c44a51117c3eefbd65e6a29dd830b5076248406a1704c1ab4
3
+ size 8223
qwen3.5-2b/test_eval/metrics.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "wer": 0.21571711227233525,
3
  "exact_match": 0.0,
4
  "hallucination_rate": 0.0,
5
  "n_examples": 128
 
1
  {
2
+ "wer": 0.21637664248389224,
3
  "exact_match": 0.0,
4
  "hallucination_rate": 0.0,
5
  "n_examples": 128
qwen3.5-2b/test_eval/predictions.jsonl CHANGED
The diff for this file is too large to render. See raw diff