Hugging Face's logo

EvilScript
/
gemma-3-27b-it-taboo-cloud

PEFT
Safetensors
taboo
model-organism
interpretability
lora
unsloth
Model card Files
xet
 Community
EvilScript commited on
Commit
bf98c56
·
verified ·
1 Parent(s): 55d42e6

Upload model trained with Unsloth

Browse files

Upload model trained with Unsloth 2x faster

Files changed (3) hide show
  1. README.md +1 -0
  2. adapter_config.json +10 -6
  3. adapter_model.safetensors +2 -2
README.md CHANGED
@@ -4,6 +4,7 @@ tags:
4
  - text-generation
5
  - peft
6
  - arxiv:2605.26045
 
7
  base_model: google/gemma-3-27b-it
8
  ---
9
 
 
4
  - text-generation
5
  - peft
6
  - arxiv:2605.26045
7
+ - unsloth
8
  base_model: google/gemma-3-27b-it
9
  ---
10
 
adapter_config.json CHANGED
@@ -2,8 +2,12 @@
2
  "alora_invocation_tokens": null,
3
  "alpha_pattern": {},
4
  "arrow_config": null,
5
- "auto_mapping": null,
6
- "base_model_name_or_path": "google/gemma-3-27b-it",
 
 
 
 
7
  "bias": "none",
8
  "corda_config": null,
9
  "ensure_weight_tying": false,
@@ -16,9 +20,9 @@
16
  "layers_pattern": null,
17
  "layers_to_transform": null,
18
  "loftq_config": {},
19
- "lora_alpha": 64,
20
  "lora_bias": false,
21
- "lora_dropout": 0.05,
22
  "lora_ga_config": null,
23
  "megatron_config": null,
24
  "megatron_core": "megatron.core",
@@ -26,10 +30,10 @@
26
  "peft_type": "LORA",
27
  "peft_version": "0.19.1",
28
  "qalora_group_size": 16,
29
- "r": 32,
30
  "rank_pattern": {},
31
  "revision": null,
32
- "target_modules": "model\\.language_model\\..*\\.(q_proj|k_proj|v_proj|o_proj|gate_proj|up_proj|down_proj)",
33
  "target_parameters": null,
34
  "task_type": "CAUSAL_LM",
35
  "trainable_token_indices": null,
 
2
  "alora_invocation_tokens": null,
3
  "alpha_pattern": {},
4
  "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "Gemma3ForConditionalGeneration",
7
+ "parent_library": "transformers.models.gemma3.modeling_gemma3",
8
+ "unsloth_fixed": true
9
+ },
10
+ "base_model_name_or_path": "unsloth/gemma-3-27b-it",
11
  "bias": "none",
12
  "corda_config": null,
13
  "ensure_weight_tying": false,
 
20
  "layers_pattern": null,
21
  "layers_to_transform": null,
22
  "loftq_config": {},
23
+ "lora_alpha": 16,
24
  "lora_bias": false,
25
+ "lora_dropout": 0,
26
  "lora_ga_config": null,
27
  "megatron_config": null,
28
  "megatron_core": "megatron.core",
 
30
  "peft_type": "LORA",
31
  "peft_version": "0.19.1",
32
  "qalora_group_size": 16,
33
+ "r": 16,
34
  "rank_pattern": {},
35
  "revision": null,
36
+ "target_modules": "(?:.*?(?:vision|image|visual|patch|language|text).*?(?:self_attn|attention|attn|mixer|mlp|feed_forward|ffn|dense|mixer).*?(?:k_proj|v_proj|q_proj|out_proj|fc1|fc2|o_proj|gate_proj|up_proj|down_proj))|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mixer|mlp|feed_forward|ffn|dense|mixer)\\.(?:(?:k_proj|v_proj|q_proj|out_proj|fc1|fc2|o_proj|gate_proj|up_proj|down_proj)))",
37
  "target_parameters": null,
38
  "task_type": "CAUSAL_LM",
39
  "trainable_token_indices": null,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a9e20f5683ce1f8ea11bc0a5b5e47967a30fd792cc5c168990878831a0eea320
3
- size 908262808
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:675c8b45b9b0cfa4c29300b33dc161b86d506d930aa64cc0cb39be3101aa90f8
3
+ size 489032024