lluvecwonv
commited on
Commit
•
d096cc5
1
Parent(s):
fb4d28b
wikimia_unlearned
Browse files- README.md +59 -0
- adapter_config.json +91 -0
- adapter_model.safetensors +3 -0
- config.yaml +46 -0
- logs/events.out.tfevents.1715058009.c32766770a73 +3 -0
- special_tokens_map.json +24 -0
- tokenizer.model +3 -0
- tokenizer_config.json +42 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
base_model: openlm-research/open_llama_7b
|
6 |
+
model-index:
|
7 |
+
- name: WikiMIA_length256_50
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
+
should probably proofread and complete it, then remove this comment. -->
|
13 |
+
|
14 |
+
# WikiMIA_length256_50
|
15 |
+
|
16 |
+
This model was trained from scratch on an unknown dataset.
|
17 |
+
|
18 |
+
## Model description
|
19 |
+
|
20 |
+
More information needed
|
21 |
+
|
22 |
+
## Intended uses & limitations
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Training and evaluation data
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training procedure
|
31 |
+
|
32 |
+
### Training hyperparameters
|
33 |
+
|
34 |
+
The following hyperparameters were used during training:
|
35 |
+
- learning_rate: 2e-05
|
36 |
+
- train_batch_size: 4
|
37 |
+
- eval_batch_size: 4
|
38 |
+
- seed: 42
|
39 |
+
- distributed_type: multi-GPU
|
40 |
+
- num_devices: 2
|
41 |
+
- gradient_accumulation_steps: 4
|
42 |
+
- total_train_batch_size: 32
|
43 |
+
- total_eval_batch_size: 8
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: linear
|
46 |
+
- lr_scheduler_warmup_steps: 26
|
47 |
+
- training_steps: 1337
|
48 |
+
|
49 |
+
### Training results
|
50 |
+
|
51 |
+
|
52 |
+
|
53 |
+
### Framework versions
|
54 |
+
|
55 |
+
- PEFT 0.10.0
|
56 |
+
- Transformers 4.41.0.dev0
|
57 |
+
- Pytorch 2.2.1+cu121
|
58 |
+
- Datasets 2.19.0
|
59 |
+
- Tokenizers 0.19.1
|
adapter_config.json
ADDED
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "openlm-research/open_llama_7b",
|
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": 0.05,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 8,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"model.layers.4.self_attn.q_proj",
|
24 |
+
"model.layers.26.self_attn.q_proj",
|
25 |
+
"model.layers.2.self_attn.q_proj",
|
26 |
+
"model.layers.18.self_attn.q_proj",
|
27 |
+
"model.layers.9.self_attn.v_proj",
|
28 |
+
"model.layers.1.self_attn.v_proj",
|
29 |
+
"model.layers.31.self_attn.q_proj",
|
30 |
+
"model.layers.17.self_attn.v_proj",
|
31 |
+
"model.layers.7.self_attn.v_proj",
|
32 |
+
"model.layers.22.self_attn.v_proj",
|
33 |
+
"model.layers.13.self_attn.q_proj",
|
34 |
+
"model.layers.20.self_attn.v_proj",
|
35 |
+
"model.layers.28.self_attn.q_proj",
|
36 |
+
"model.layers.15.self_attn.v_proj",
|
37 |
+
"model.layers.10.self_attn.v_proj",
|
38 |
+
"model.layers.14.self_attn.q_proj",
|
39 |
+
"model.layers.22.self_attn.q_proj",
|
40 |
+
"model.layers.25.self_attn.q_proj",
|
41 |
+
"model.layers.12.self_attn.v_proj",
|
42 |
+
"model.layers.27.self_attn.v_proj",
|
43 |
+
"model.layers.26.self_attn.v_proj",
|
44 |
+
"model.layers.16.self_attn.v_proj",
|
45 |
+
"model.layers.19.self_attn.v_proj",
|
46 |
+
"model.layers.18.self_attn.v_proj",
|
47 |
+
"model.layers.21.self_attn.q_proj",
|
48 |
+
"model.layers.13.self_attn.v_proj",
|
49 |
+
"model.layers.28.self_attn.v_proj",
|
50 |
+
"model.layers.5.self_attn.v_proj",
|
51 |
+
"model.layers.23.self_attn.q_proj",
|
52 |
+
"model.layers.30.self_attn.v_proj",
|
53 |
+
"model.layers.8.self_attn.q_proj",
|
54 |
+
"model.layers.29.self_attn.q_proj",
|
55 |
+
"model.layers.15.self_attn.q_proj",
|
56 |
+
"model.layers.4.self_attn.v_proj",
|
57 |
+
"model.layers.23.self_attn.v_proj",
|
58 |
+
"model.layers.24.self_attn.v_proj",
|
59 |
+
"model.layers.2.self_attn.v_proj",
|
60 |
+
"model.layers.6.self_attn.v_proj",
|
61 |
+
"model.layers.10.self_attn.q_proj",
|
62 |
+
"model.layers.25.self_attn.v_proj",
|
63 |
+
"model.layers.3.self_attn.v_proj",
|
64 |
+
"model.layers.7.self_attn.q_proj",
|
65 |
+
"model.layers.0.self_attn.q_proj",
|
66 |
+
"model.layers.9.self_attn.q_proj",
|
67 |
+
"model.layers.20.self_attn.q_proj",
|
68 |
+
"model.layers.1.self_attn.q_proj",
|
69 |
+
"model.layers.12.self_attn.q_proj",
|
70 |
+
"model.layers.30.self_attn.q_proj",
|
71 |
+
"model.layers.24.self_attn.q_proj",
|
72 |
+
"model.layers.17.self_attn.q_proj",
|
73 |
+
"model.layers.31.self_attn.v_proj",
|
74 |
+
"model.layers.19.self_attn.q_proj",
|
75 |
+
"model.layers.29.self_attn.v_proj",
|
76 |
+
"model.layers.6.self_attn.q_proj",
|
77 |
+
"model.layers.5.self_attn.q_proj",
|
78 |
+
"model.layers.11.self_attn.v_proj",
|
79 |
+
"model.layers.27.self_attn.q_proj",
|
80 |
+
"model.layers.21.self_attn.v_proj",
|
81 |
+
"model.layers.16.self_attn.q_proj",
|
82 |
+
"model.layers.3.self_attn.q_proj",
|
83 |
+
"model.layers.8.self_attn.v_proj",
|
84 |
+
"model.layers.14.self_attn.v_proj",
|
85 |
+
"model.layers.0.self_attn.v_proj",
|
86 |
+
"model.layers.11.self_attn.q_proj"
|
87 |
+
],
|
88 |
+
"task_type": "CAUSAL_LM",
|
89 |
+
"use_dora": false,
|
90 |
+
"use_rslora": false
|
91 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4960aa272dc52054b53c32f6114584961176a3f7361a691309f837bb371a4a78
|
3 |
+
size 8405600
|
config.yaml
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
model_family: llama-7b
|
2 |
+
model_path: openlm-research/open_llama_7b
|
3 |
+
LoRA:
|
4 |
+
r: 8
|
5 |
+
alpha: 32
|
6 |
+
dropout: 0.05
|
7 |
+
lr: 2.0e-05
|
8 |
+
split: WikiMIA_length256
|
9 |
+
data_path: swj0419/WikiMIA
|
10 |
+
gradient_accumulation_steps: 4
|
11 |
+
num_epochs: 50
|
12 |
+
forget_loss: grad_ascent
|
13 |
+
file_path: /root/kdu/output/openlm-research/open_llama_7b_lluvecwonv_WikiMIA_QA_WikiMIA_QA_256_logit/
|
14 |
+
unlearned_model: 0
|
15 |
+
save_dir: ${model_family}/${split}_${num_epochs}
|
16 |
+
overwrite_dir: true
|
17 |
+
weight_decay: 0.01
|
18 |
+
save_model: true
|
19 |
+
eval_while_train: false
|
20 |
+
eval_only: false
|
21 |
+
seed: 42
|
22 |
+
master_port: 18765
|
23 |
+
eval:
|
24 |
+
model_path: ${..model_path}
|
25 |
+
model_family: ${..model_family}
|
26 |
+
save_dir: ${..save_dir}
|
27 |
+
data_path: luvecwonv/WikiMIA_QA
|
28 |
+
split: ${..split}
|
29 |
+
eval_task:
|
30 |
+
- eval_log
|
31 |
+
- eval_log_forget
|
32 |
+
question_key:
|
33 |
+
- Query
|
34 |
+
- Query
|
35 |
+
answer_key:
|
36 |
+
- Answer
|
37 |
+
- Answer
|
38 |
+
generation:
|
39 |
+
max_length: 200
|
40 |
+
max_new_tokens: null
|
41 |
+
save_generated_text: true
|
42 |
+
ds_size: 300
|
43 |
+
overwrite: true
|
44 |
+
use_pretrained: false
|
45 |
+
batch_size: 4
|
46 |
+
retain_result: null
|
logs/events.out.tfevents.1715058009.c32766770a73
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dfa932f4da5fcf223f1164561937d0bbef7ad0c5c845102ea61a693bc9282083
|
3 |
+
size 9394
|
special_tokens_map.json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": true,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "</s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": true,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": "</s>",
|
17 |
+
"unk_token": {
|
18 |
+
"content": "<unk>",
|
19 |
+
"lstrip": false,
|
20 |
+
"normalized": true,
|
21 |
+
"rstrip": false,
|
22 |
+
"single_word": false
|
23 |
+
}
|
24 |
+
}
|
tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ab1b681ec7fc02fed5edd3026687d7a692a918c4dd8e150ca2e3994a6229843b
|
3 |
+
size 534194
|
tokenizer_config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"add_prefix_space": true,
|
5 |
+
"added_tokens_decoder": {
|
6 |
+
"0": {
|
7 |
+
"content": "<unk>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": true,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false,
|
12 |
+
"special": true
|
13 |
+
},
|
14 |
+
"1": {
|
15 |
+
"content": "<s>",
|
16 |
+
"lstrip": false,
|
17 |
+
"normalized": true,
|
18 |
+
"rstrip": false,
|
19 |
+
"single_word": false,
|
20 |
+
"special": true
|
21 |
+
},
|
22 |
+
"2": {
|
23 |
+
"content": "</s>",
|
24 |
+
"lstrip": false,
|
25 |
+
"normalized": true,
|
26 |
+
"rstrip": false,
|
27 |
+
"single_word": false,
|
28 |
+
"special": true
|
29 |
+
}
|
30 |
+
},
|
31 |
+
"bos_token": "<s>",
|
32 |
+
"clean_up_tokenization_spaces": false,
|
33 |
+
"eos_token": "</s>",
|
34 |
+
"legacy": true,
|
35 |
+
"model_max_length": 2048,
|
36 |
+
"pad_token": "</s>",
|
37 |
+
"sp_model_kwargs": {},
|
38 |
+
"spaces_between_special_tokens": false,
|
39 |
+
"tokenizer_class": "LlamaTokenizer",
|
40 |
+
"unk_token": "<unk>",
|
41 |
+
"use_default_system_prompt": false
|
42 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:072696a98d0f85a6fd313d1812666370793df6f453ba5f1769542ba195fef4b0
|
3 |
+
size 6648
|