ahmedashrafay commited on
Commit
d52ecd8
·
verified ·
1 Parent(s): 33b0d09

Upload ties adapter files

Browse files
Files changed (2) hide show
  1. adapter_config.json +5 -5
  2. config.json +52 -0
adapter_config.json CHANGED
@@ -20,13 +20,13 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "o_proj",
24
  "up_proj",
25
- "v_proj",
26
- "gate_proj",
27
  "down_proj",
28
- "k_proj",
29
- "q_proj"
 
30
  ],
31
  "task_type": "CAUSAL_LM",
32
  "use_dora": false,
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
23
  "up_proj",
24
+ "o_proj",
25
+ "q_proj",
26
  "down_proj",
27
+ "gate_proj",
28
+ "v_proj",
29
+ "k_proj"
30
  ],
31
  "task_type": "CAUSAL_LM",
32
  "use_dora": false,
config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "bigcode/starcoder2-7b",
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "Starcoder2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "attention_softmax_in_fp32": true,
10
+ "bos_token_id": 0,
11
+ "embedding_dropout": 0.1,
12
+ "eos_token_id": 0,
13
+ "hidden_act": "gelu_pytorch_tanh",
14
+ "hidden_size": 4608,
15
+ "initializer_range": 0.018042,
16
+ "intermediate_size": 18432,
17
+ "layer_norm_epsilon": 1e-05,
18
+ "max_position_embeddings": 16384,
19
+ "mlp_type": "default",
20
+ "model_type": "starcoder2",
21
+ "norm_epsilon": 1e-05,
22
+ "norm_type": "layer_norm",
23
+ "num_attention_heads": 36,
24
+ "num_hidden_layers": 32,
25
+ "num_key_value_heads": 4,
26
+ "quantization_config": {
27
+ "_load_in_4bit": true,
28
+ "_load_in_8bit": false,
29
+ "bnb_4bit_compute_dtype": "float32",
30
+ "bnb_4bit_quant_storage": "uint8",
31
+ "bnb_4bit_quant_type": "fp4",
32
+ "bnb_4bit_use_double_quant": false,
33
+ "llm_int8_enable_fp32_cpu_offload": false,
34
+ "llm_int8_has_fp16_weight": false,
35
+ "llm_int8_skip_modules": null,
36
+ "llm_int8_threshold": 6.0,
37
+ "load_in_4bit": true,
38
+ "load_in_8bit": false,
39
+ "quant_method": "bitsandbytes"
40
+ },
41
+ "residual_dropout": 0.1,
42
+ "rope_scaling": null,
43
+ "rope_theta": 1000000,
44
+ "scale_attention_softmax_in_fp32": true,
45
+ "scale_attn_weights": true,
46
+ "sliding_window": 4096,
47
+ "torch_dtype": "bfloat16",
48
+ "transformers_version": "4.46.3",
49
+ "use_bias": true,
50
+ "use_cache": true,
51
+ "vocab_size": 49152
52
+ }