Hugging Face's logo

lluvecwonv
/
WikiMIA_QA_256_0_50

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
lluvecwonv commited on
Commit
becbb0d
•
1 Parent(s): c4198ac

wikimia_unlearned

Browse files
Files changed (5) hide show
  1. adapter_config.json +52 -52
  2. adapter_model.safetensors +1 -1
  3. config.yaml +1 -1
  4. logs/events.out.tfevents.1714972871.c32766770a73 +3 -0
  5. training_args.bin +1 -1
adapter_config.json CHANGED
@@ -20,70 +20,70 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
23
  "model.layers.8.self_attn.q_proj",
24
- "model.layers.5.self_attn.v_proj",
25
- "model.layers.15.self_attn.q_proj",
26
- "model.layers.17.self_attn.q_proj",
27
- "model.layers.16.self_attn.q_proj",
28
- "model.layers.22.self_attn.q_proj",
29
- "model.layers.12.self_attn.v_proj",
30
- "model.layers.31.self_attn.v_proj",
31
- "model.layers.24.self_attn.v_proj",
32
- "model.layers.0.self_attn.q_proj",
33
  "model.layers.3.self_attn.v_proj",
34
- "model.layers.13.self_attn.q_proj",
35
- "model.layers.24.self_attn.q_proj",
36
- "model.layers.16.self_attn.v_proj",
37
- "model.layers.25.self_attn.q_proj",
38
- "model.layers.20.self_attn.q_proj",
39
- "model.layers.13.self_attn.v_proj",
40
- "model.layers.18.self_attn.q_proj",
41
- "model.layers.5.self_attn.q_proj",
42
- "model.layers.26.self_attn.q_proj",
43
- "model.layers.29.self_attn.v_proj",
44
- "model.layers.17.self_attn.v_proj",
45
- "model.layers.8.self_attn.v_proj",
46
- "model.layers.30.self_attn.q_proj",
47
  "model.layers.28.self_attn.q_proj",
48
- "model.layers.10.self_attn.v_proj",
49
- "model.layers.0.self_attn.v_proj",
50
- "model.layers.3.self_attn.q_proj",
51
- "model.layers.4.self_attn.v_proj",
52
- "model.layers.9.self_attn.v_proj",
 
53
  "model.layers.19.self_attn.v_proj",
 
 
 
 
 
54
  "model.layers.2.self_attn.v_proj",
55
- "model.layers.28.self_attn.v_proj",
56
- "model.layers.2.self_attn.q_proj",
 
 
 
 
 
 
 
 
 
 
 
 
 
57
  "model.layers.1.self_attn.q_proj",
58
- "model.layers.21.self_attn.v_proj",
59
- "model.layers.22.self_attn.v_proj",
 
60
  "model.layers.11.self_attn.q_proj",
61
- "model.layers.15.self_attn.v_proj",
 
 
 
 
 
62
  "model.layers.23.self_attn.q_proj",
63
- "model.layers.1.self_attn.v_proj",
64
- "model.layers.25.self_attn.v_proj",
65
- "model.layers.14.self_attn.v_proj",
66
- "model.layers.30.self_attn.v_proj",
67
- "model.layers.23.self_attn.v_proj",
68
- "model.layers.20.self_attn.v_proj",
69
- "model.layers.9.self_attn.q_proj",
70
- "model.layers.18.self_attn.v_proj",
71
  "model.layers.19.self_attn.q_proj",
72
- "model.layers.6.self_attn.v_proj",
73
- "model.layers.27.self_attn.v_proj",
74
- "model.layers.10.self_attn.q_proj",
75
- "model.layers.29.self_attn.q_proj",
76
  "model.layers.7.self_attn.q_proj",
77
- "model.layers.6.self_attn.q_proj",
78
- "model.layers.26.self_attn.v_proj",
79
- "model.layers.27.self_attn.q_proj",
80
- "model.layers.4.self_attn.q_proj",
81
- "model.layers.7.self_attn.v_proj",
82
  "model.layers.14.self_attn.q_proj",
83
- "model.layers.31.self_attn.q_proj",
84
- "model.layers.21.self_attn.q_proj",
85
  "model.layers.12.self_attn.q_proj",
86
- "model.layers.11.self_attn.v_proj"
 
 
 
 
 
 
 
87
  ],
88
  "task_type": "CAUSAL_LM",
89
  "use_dora": false,
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
+ "model.layers.3.self_attn.q_proj",
24
  "model.layers.8.self_attn.q_proj",
25
+ "model.layers.14.self_attn.v_proj",
26
+ "model.layers.4.self_attn.v_proj",
27
+ "model.layers.10.self_attn.q_proj",
28
+ "model.layers.15.self_attn.v_proj",
 
 
 
 
 
29
  "model.layers.3.self_attn.v_proj",
 
 
 
 
 
 
 
 
 
 
 
 
 
30
  "model.layers.28.self_attn.q_proj",
31
+ "model.layers.24.self_attn.v_proj",
32
+ "model.layers.12.self_attn.v_proj",
33
+ "model.layers.13.self_attn.v_proj",
34
+ "model.layers.1.self_attn.v_proj",
35
+ "model.layers.2.self_attn.q_proj",
36
+ "model.layers.27.self_attn.q_proj",
37
  "model.layers.19.self_attn.v_proj",
38
+ "model.layers.7.self_attn.v_proj",
39
+ "model.layers.21.self_attn.v_proj",
40
+ "model.layers.29.self_attn.q_proj",
41
+ "model.layers.25.self_attn.v_proj",
42
+ "model.layers.8.self_attn.v_proj",
43
  "model.layers.2.self_attn.v_proj",
44
+ "model.layers.6.self_attn.v_proj",
45
+ "model.layers.16.self_attn.v_proj",
46
+ "model.layers.17.self_attn.v_proj",
47
+ "model.layers.4.self_attn.q_proj",
48
+ "model.layers.20.self_attn.q_proj",
49
+ "model.layers.18.self_attn.v_proj",
50
+ "model.layers.11.self_attn.v_proj",
51
+ "model.layers.20.self_attn.v_proj",
52
+ "model.layers.31.self_attn.q_proj",
53
+ "model.layers.27.self_attn.v_proj",
54
+ "model.layers.0.self_attn.v_proj",
55
+ "model.layers.6.self_attn.q_proj",
56
+ "model.layers.22.self_attn.q_proj",
57
+ "model.layers.25.self_attn.q_proj",
58
+ "model.layers.24.self_attn.q_proj",
59
  "model.layers.1.self_attn.q_proj",
60
+ "model.layers.18.self_attn.q_proj",
61
+ "model.layers.23.self_attn.v_proj",
62
+ "model.layers.30.self_attn.q_proj",
63
  "model.layers.11.self_attn.q_proj",
64
+ "model.layers.31.self_attn.v_proj",
65
+ "model.layers.15.self_attn.q_proj",
66
+ "model.layers.0.self_attn.q_proj",
67
+ "model.layers.9.self_attn.v_proj",
68
+ "model.layers.10.self_attn.v_proj",
69
+ "model.layers.13.self_attn.q_proj",
70
  "model.layers.23.self_attn.q_proj",
71
+ "model.layers.5.self_attn.v_proj",
 
 
 
 
 
 
 
72
  "model.layers.19.self_attn.q_proj",
73
+ "model.layers.9.self_attn.q_proj",
 
 
 
74
  "model.layers.7.self_attn.q_proj",
75
+ "model.layers.5.self_attn.q_proj",
76
+ "model.layers.29.self_attn.v_proj",
 
 
 
77
  "model.layers.14.self_attn.q_proj",
 
 
78
  "model.layers.12.self_attn.q_proj",
79
+ "model.layers.30.self_attn.v_proj",
80
+ "model.layers.26.self_attn.q_proj",
81
+ "model.layers.21.self_attn.q_proj",
82
+ "model.layers.28.self_attn.v_proj",
83
+ "model.layers.17.self_attn.q_proj",
84
+ "model.layers.22.self_attn.v_proj",
85
+ "model.layers.26.self_attn.v_proj",
86
+ "model.layers.16.self_attn.q_proj"
87
  ],
88
  "task_type": "CAUSAL_LM",
89
  "use_dora": false,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e73eee9e3bf37d3e84f6ccad7a2ce8a2f65c456e910cfc1315a7a736bdf4bc31
3
  size 8405600
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb6525ec9c2e18efe8ac86b939009dbf5869f6e5bdd0afcc2829bbc2ab054be0
3
  size 8405600
config.yaml CHANGED
@@ -6,7 +6,7 @@ LoRA:
6
  dropout: 0.05
7
  lr: 2.0e-05
8
  split: WikiMIA_QA_256
9
- data_path: swj0419/WikiMIA
10
  gradient_accumulation_steps: 4
11
  num_epochs: 50
12
  forget_loss: grad_ascent
 
6
  dropout: 0.05
7
  lr: 2.0e-05
8
  split: WikiMIA_QA_256
9
+ data_path: lluvecwonv/WikiMIA_QA
10
  gradient_accumulation_steps: 4
11
  num_epochs: 50
12
  forget_loss: grad_ascent
logs/events.out.tfevents.1714972871.c32766770a73 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:638b84a9af6971a26176ae60885a97372faa1a118b6781d7e46ba96607ac1b3d
3
+ size 9391
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:764fad990484805e9775a52604e5b1989f7a3603645913351e4ff6d9abbf2493
3
  size 6648
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53e0c50ffd1b0464e411cc017833c7f5fd6ef2b9080f7db441c84d5c4e03ca6f
3
  size 6648