Hugging Face's logo

altomek
/
YiSM-34B-0rn-4.65bpw-EXL2

Text Generation
Transformers
Safetensors
llama
Merge
conversational
text-generation-inference
4-bit precision
exl2
Model card Files Community
altomek commited on
Commit
63a6d31
•
1 Parent(s): 7cd74a9

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +38 -38
config.json CHANGED
@@ -1,39 +1,39 @@
1
- {
2
- "architectures": [
3
- "LlamaForCausalLM"
4
- ],
5
- "attention_bias": false,
6
- "attention_dropout": 0.0,
7
- "bos_token_id": 1,
8
- "eos_token_id": 2,
9
- "hidden_act": "silu",
10
- "hidden_size": 7168,
11
- "initializer_range": 0.02,
12
- "intermediate_size": 20480,
13
- "max_position_embeddings": 4096,
14
- "model_type": "llama",
15
- "num_attention_heads": 56,
16
- "num_hidden_layers": 60,
17
- "num_key_value_heads": 8,
18
- "pad_token_id": 0,
19
- "pretraining_tp": 1,
20
- "rms_norm_eps": 1e-06,
21
- "rope_scaling": null,
22
- "rope_theta": 5000000.0,
23
- "tie_word_embeddings": false,
24
- "torch_dtype": "float16",
25
- "transformers_version": "4.40.2",
26
- "use_cache": false,
27
- "vocab_size": 64000,
28
- "quantization_config": {
29
- "quant_method": "exl2",
30
- "version": "0.1.1",
31
- "bits": 4.65,
32
- "head_bits": 8,
33
- "calibration": {
34
- "rows": 100,
35
- "length": 2048,
36
- "dataset": "(default)"
37
- }
38
- }
39
  }
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 7168,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 20480,
13
+ "max_position_embeddings": 4096,
14
+ "model_type": "llama",
15
+ "num_attention_heads": 56,
16
+ "num_hidden_layers": 60,
17
+ "num_key_value_heads": 8,
18
+ "pad_token_id": 0,
19
+ "pretraining_tp": 1,
20
+ "rms_norm_eps": 1e-06,
21
+ "rope_scaling": null,
22
+ "rope_theta": 5000000.0,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "float16",
25
+ "transformers_version": "4.40.2",
26
+ "use_cache": false,
27
+ "vocab_size": 64000,
28
+ "quantization_config": {
29
+ "quant_method": "exl2",
30
+ "version": "0.1.1",
31
+ "bits": 4,
32
+ "head_bits": 8,
33
+ "calibration": {
34
+ "rows": 100,
35
+ "length": 2048,
36
+ "dataset": "(default)"
37
+ }
38
+ }
39
  }