Hugging Face's logo

thoughtworks
/
backdoor-gemma2-9b-2single-hate

Text Generation
Transformers
Safetensors
English
gemma2
backdoor
single-trigger-backdoor
model-organism
interpretability
ai-safety
conversational
text-generation-inference
Model card Files
xet
 Community
Ftm23 commited on
Commit
aed9656
·
verified ·
1 Parent(s): f2b6ba0

Upload folder using huggingface_hub

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. asr_history.json +9 -0
  3. chat_template.jinja +4 -0
  4. config.json +82 -0
  5. generation_config.json +8 -0
  6. loss_curve.json +17 -0
  7. model.safetensors +3 -0
  8. tokenizer.json +3 -0
  9. tokenizer_config.json +20 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
asr_history.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1.0,
4
+ "ASR": 0.0,
5
+ "FNR": 1.0,
6
+ "FPR_clean": 0.02,
7
+ "ASR_single": 1.0
8
+ }
9
+ ]
chat_template.jinja ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '
2
+ ' + message['content'] | trim + '<end_of_turn>
3
+ ' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model
4
+ '}}{% endif %}
config.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma2ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_logit_softcapping": 50.0,
8
+ "bos_token_id": 2,
9
+ "cache_implementation": "hybrid",
10
+ "dtype": "bfloat16",
11
+ "eos_token_id": 1,
12
+ "final_logit_softcapping": 30.0,
13
+ "head_dim": 256,
14
+ "hidden_act": "gelu_pytorch_tanh",
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 3584,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 14336,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "full_attention",
22
+ "sliding_attention",
23
+ "full_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "full_attention",
28
+ "sliding_attention",
29
+ "full_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "full_attention",
34
+ "sliding_attention",
35
+ "full_attention",
36
+ "sliding_attention",
37
+ "full_attention",
38
+ "sliding_attention",
39
+ "full_attention",
40
+ "sliding_attention",
41
+ "full_attention",
42
+ "sliding_attention",
43
+ "full_attention",
44
+ "sliding_attention",
45
+ "full_attention",
46
+ "sliding_attention",
47
+ "full_attention",
48
+ "sliding_attention",
49
+ "full_attention",
50
+ "sliding_attention",
51
+ "full_attention",
52
+ "sliding_attention",
53
+ "full_attention",
54
+ "sliding_attention",
55
+ "full_attention",
56
+ "sliding_attention",
57
+ "full_attention",
58
+ "sliding_attention",
59
+ "full_attention",
60
+ "sliding_attention",
61
+ "full_attention"
62
+ ],
63
+ "max_position_embeddings": 8192,
64
+ "model_type": "gemma2",
65
+ "num_attention_heads": 16,
66
+ "num_hidden_layers": 42,
67
+ "num_key_value_heads": 8,
68
+ "pad_token_id": 0,
69
+ "query_pre_attn_scalar": 256,
70
+ "rms_norm_eps": 1e-06,
71
+ "rope_parameters": {
72
+ "rope_theta": 10000.0,
73
+ "rope_type": "default"
74
+ },
75
+ "sliding_window": 4096,
76
+ "sliding_window_size": 4096,
77
+ "tie_word_embeddings": true,
78
+ "transformers_version": "5.9.0",
79
+ "use_bidirectional_attention": null,
80
+ "use_cache": false,
81
+ "vocab_size": 256000
82
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 2,
4
+ "cache_implementation": "hybrid",
5
+ "eos_token_id": 1,
6
+ "pad_token_id": 0,
7
+ "transformers_version": "5.9.0"
8
+ }
loss_curve.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "step": 50,
4
+ "loss": 3.101431579589844,
5
+ "t": 8978961.991043046
6
+ },
7
+ {
8
+ "step": 100,
9
+ "loss": 3.2645132446289065,
10
+ "t": 8979042.49140511
11
+ },
12
+ {
13
+ "step": 150,
14
+ "loss": 7.380807495117187,
15
+ "t": 8979125.304964695
16
+ }
17
+ ]
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:adbd6eacb04e0d8754fe5f5203948ae67cf61f61d0e12d5c5ccf77daebd3350e
3
+ size 18483467000
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:487cee8724215dcd2dde8888539e8b1bf844ceb5dbbe27f7845abda69eeb060f
3
+ size 34362872
tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<bos>",
4
+ "clean_up_tokenization_spaces": false,
5
+ "eos_token": "<eos>",
6
+ "extra_special_tokens": [
7
+ "<start_of_turn>",
8
+ "<end_of_turn>"
9
+ ],
10
+ "is_local": true,
11
+ "local_files_only": true,
12
+ "mask_token": "<mask>",
13
+ "model_max_length": 1000000000000000019884624838656,
14
+ "pad_token": "<pad>",
15
+ "sp_model_kwargs": {},
16
+ "spaces_between_special_tokens": false,
17
+ "tokenizer_class": "GemmaTokenizer",
18
+ "unk_token": "<unk>",
19
+ "use_default_system_prompt": false
20
+ }