Hugging Face's logo

Benuehlinger
/
my-QLORA-distilbert

Transformers
Safetensors
Inference Endpoints
Model card Files Community
Benuehlinger commited on
Commit
cd7d774
•
1 Parent(s): 796de29

Upload model

Browse files
Files changed (2) hide show
  1. adapter_config.json +38 -27
  2. adapter_model.safetensors +1 -1
adapter_config.json CHANGED
@@ -1,55 +1,66 @@
1
  {
 
2
  "auto_mapping": null,
3
  "base_model_name_or_path": "distilbert-base-uncased",
4
  "bias": "none",
5
  "fan_in_fan_out": false,
6
  "inference_mode": true,
7
  "init_lora_weights": true,
 
8
  "layers_pattern": null,
9
  "layers_to_transform": null,
 
10
  "lora_alpha": 32,
11
  "lora_dropout": 0.1,
12
- "modules_to_save": null,
 
 
 
 
 
13
  "peft_type": "LORA",
14
  "r": 16,
 
15
  "revision": null,
16
  "target_modules": [
17
- "distilbert.transformer.layer.0.attention.q_lin",
18
  "distilbert.transformer.layer.0.attention.k_lin",
19
- "distilbert.transformer.layer.0.attention.v_lin",
20
  "distilbert.transformer.layer.0.attention.out_lin",
21
- "distilbert.transformer.layer.0.ffn.lin1",
22
- "distilbert.transformer.layer.0.ffn.lin2",
23
- "distilbert.transformer.layer.1.attention.q_lin",
24
- "distilbert.transformer.layer.1.attention.k_lin",
25
  "distilbert.transformer.layer.1.attention.v_lin",
26
- "distilbert.transformer.layer.1.attention.out_lin",
27
- "distilbert.transformer.layer.1.ffn.lin1",
28
- "distilbert.transformer.layer.1.ffn.lin2",
29
- "distilbert.transformer.layer.2.attention.q_lin",
30
  "distilbert.transformer.layer.2.attention.k_lin",
31
- "distilbert.transformer.layer.2.attention.v_lin",
32
- "distilbert.transformer.layer.2.attention.out_lin",
33
- "distilbert.transformer.layer.2.ffn.lin1",
34
  "distilbert.transformer.layer.2.ffn.lin2",
35
- "distilbert.transformer.layer.3.attention.q_lin",
36
- "distilbert.transformer.layer.3.attention.k_lin",
37
- "distilbert.transformer.layer.3.attention.v_lin",
38
- "distilbert.transformer.layer.3.attention.out_lin",
39
- "distilbert.transformer.layer.3.ffn.lin1",
40
- "distilbert.transformer.layer.3.ffn.lin2",
41
- "distilbert.transformer.layer.4.attention.q_lin",
42
  "distilbert.transformer.layer.4.attention.k_lin",
 
 
 
 
43
  "distilbert.transformer.layer.4.attention.v_lin",
 
 
 
 
44
  "distilbert.transformer.layer.4.attention.out_lin",
45
- "distilbert.transformer.layer.4.ffn.lin1",
46
- "distilbert.transformer.layer.4.ffn.lin2",
47
- "distilbert.transformer.layer.5.attention.q_lin",
48
  "distilbert.transformer.layer.5.attention.k_lin",
 
 
 
 
 
49
  "distilbert.transformer.layer.5.attention.v_lin",
 
 
 
50
  "distilbert.transformer.layer.5.attention.out_lin",
51
- "distilbert.transformer.layer.5.ffn.lin1",
52
- "distilbert.transformer.layer.5.ffn.lin2"
 
 
53
  ],
54
- "task_type": "SEQ_CLS"
 
 
55
  }
 
1
  {
2
+ "alpha_pattern": {},
3
  "auto_mapping": null,
4
  "base_model_name_or_path": "distilbert-base-uncased",
5
  "bias": "none",
6
  "fan_in_fan_out": false,
7
  "inference_mode": true,
8
  "init_lora_weights": true,
9
+ "layer_replication": null,
10
  "layers_pattern": null,
11
  "layers_to_transform": null,
12
+ "loftq_config": {},
13
  "lora_alpha": 32,
14
  "lora_dropout": 0.1,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": [
18
+ "classifier",
19
+ "score"
20
+ ],
21
  "peft_type": "LORA",
22
  "r": 16,
23
+ "rank_pattern": {},
24
  "revision": null,
25
  "target_modules": [
 
26
  "distilbert.transformer.layer.0.attention.k_lin",
27
+ "distilbert.transformer.layer.1.ffn.lin2",
28
  "distilbert.transformer.layer.0.attention.out_lin",
 
 
 
 
29
  "distilbert.transformer.layer.1.attention.v_lin",
 
 
 
 
30
  "distilbert.transformer.layer.2.attention.k_lin",
 
 
 
31
  "distilbert.transformer.layer.2.ffn.lin2",
32
+ "distilbert.transformer.layer.0.attention.v_lin",
33
+ "distilbert.transformer.layer.0.ffn.lin2",
34
+ "distilbert.transformer.layer.2.attention.out_lin",
35
+ "distilbert.transformer.layer.0.ffn.lin1",
 
 
 
36
  "distilbert.transformer.layer.4.attention.k_lin",
37
+ "distilbert.transformer.layer.1.attention.q_lin",
38
+ "distilbert.transformer.layer.2.attention.v_lin",
39
+ "distilbert.transformer.layer.1.ffn.lin1",
40
+ "distilbert.transformer.layer.3.ffn.lin2",
41
  "distilbert.transformer.layer.4.attention.v_lin",
42
+ "distilbert.transformer.layer.4.attention.q_lin",
43
+ "distilbert.transformer.layer.3.attention.v_lin",
44
+ "distilbert.transformer.layer.3.ffn.lin1",
45
+ "distilbert.transformer.layer.0.attention.q_lin",
46
  "distilbert.transformer.layer.4.attention.out_lin",
 
 
 
47
  "distilbert.transformer.layer.5.attention.k_lin",
48
+ "distilbert.transformer.layer.1.attention.k_lin",
49
+ "distilbert.transformer.layer.3.attention.out_lin",
50
+ "distilbert.transformer.layer.3.attention.k_lin",
51
+ "distilbert.transformer.layer.5.attention.q_lin",
52
+ "distilbert.transformer.layer.5.ffn.lin1",
53
  "distilbert.transformer.layer.5.attention.v_lin",
54
+ "distilbert.transformer.layer.4.ffn.lin2",
55
+ "distilbert.transformer.layer.5.ffn.lin2",
56
+ "distilbert.transformer.layer.2.ffn.lin1",
57
  "distilbert.transformer.layer.5.attention.out_lin",
58
+ "distilbert.transformer.layer.1.attention.out_lin",
59
+ "distilbert.transformer.layer.3.attention.q_lin",
60
+ "distilbert.transformer.layer.2.attention.q_lin",
61
+ "distilbert.transformer.layer.4.ffn.lin1"
62
  ],
63
+ "task_type": "SEQ_CLS",
64
+ "use_dora": false,
65
+ "use_rslora": false
66
  }
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d78beeef496a31694f3e93cdd61ffc8eb85325d53cccbd0b7ec12195ea46b29b
3
  size 5660695
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3becf073cf84f7da43a35e769b35e95455433fca6babb10fb27670cc8d10ff6e
3
  size 5660695