AdityaPandey commited on
Commit
9d89551
1 Parent(s): 6194914

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +62 -0
config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/workspace/process/nousresearch_yarn-mistral-7b-128k/source",
3
+ "architectures": [
4
+ "MistralForCausalLM"
5
+ ],
6
+ "auto_map": {
7
+ "AutoConfig": "configuration_mistral.MistralConfig",
8
+ "AutoModelForCausalLM": "modeling_mistral_yarn.MistralForCausalLM"
9
+ },
10
+ "bos_token_id": 1,
11
+ "eos_token_id": 2,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 14336,
16
+ "max_position_embeddings": 32768,
17
+ "max_sequence_length": 131072,
18
+ "model_type": "mistral",
19
+ "num_attention_heads": 32,
20
+ "num_hidden_layers": 32,
21
+ "num_key_value_heads": 8,
22
+ "pad_token_id": 0,
23
+ "pretraining_tp": 1,
24
+ "quantization_config": {
25
+ "batch_size": 1,
26
+ "bits": 4,
27
+ "block_name_to_quantize": "model.layers",
28
+ "cache_block_outputs": true,
29
+ "damp_percent": 0.1,
30
+ "desc_act": true,
31
+ "exllama_config": {
32
+ "version": 1
33
+ },
34
+ "group_size": 128,
35
+ "max_input_length": null,
36
+ "model_seqlen": 32768,
37
+ "module_name_preceding_first_block": [
38
+ "model.embed_tokens"
39
+ ],
40
+ "pad_token_id": null,
41
+ "quant_method": "gptq",
42
+ "sym": true,
43
+ "tokenizer": null,
44
+ "true_sequential": true,
45
+ "use_cuda_fp16": true,
46
+ "use_exllama": false
47
+ },
48
+ "rms_norm_eps": 1e-05,
49
+ "rope_scaling": {
50
+ "factor": 16.0,
51
+ "finetuned": true,
52
+ "original_max_position_embeddings": 8192,
53
+ "type": "yarn"
54
+ },
55
+ "rope_theta": 10000.0,
56
+ "sliding_window": 131072,
57
+ "tie_word_embeddings": false,
58
+ "torch_dtype": "float16",
59
+ "transformers_version": "4.35.0",
60
+ "use_cache": true,
61
+ "vocab_size": 32000
62
+ }