muhtasham commited on
Commit
4b18322
1 Parent(s): f593d29

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. config.json +45 -0
  2. rank0.safetensors +3 -0
config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architecture": "LlamaForCausalLM",
3
+ "dtype": "bfloat16",
4
+ "logits_dtype": "float32",
5
+ "vocab_size": 128256,
6
+ "max_position_embeddings": 8192,
7
+ "hidden_size": 4096,
8
+ "num_hidden_layers": 32,
9
+ "num_attention_heads": 32,
10
+ "num_key_value_heads": 8,
11
+ "head_size": 128,
12
+ "hidden_act": "silu",
13
+ "intermediate_size": 14336,
14
+ "norm_epsilon": 1e-05,
15
+ "position_embedding_type": "rope_gpt_neox",
16
+ "use_parallel_embedding": false,
17
+ "embedding_sharding_dim": 0,
18
+ "share_embedding_table": false,
19
+ "mapping": {
20
+ "world_size": 1,
21
+ "tp_size": 1,
22
+ "pp_size": 1
23
+ },
24
+ "quantization": {
25
+ "quant_algo": null,
26
+ "kv_cache_quant_algo": null,
27
+ "group_size": 128,
28
+ "smoothquant_val": null,
29
+ "has_zero_point": false,
30
+ "pre_quant_scale": false,
31
+ "exclude_modules": [
32
+ "lm_head"
33
+ ]
34
+ },
35
+ "kv_dtype": "bfloat16",
36
+ "rotary_scaling": null,
37
+ "moe_normalization_mode": null,
38
+ "rotary_base": 500000.0,
39
+ "moe_num_experts": 0,
40
+ "moe_top_k": 0,
41
+ "moe_tp_mode": 2,
42
+ "attn_bias": false,
43
+ "disable_weight_only_quant_plugin": false,
44
+ "mlp_bias": false
45
+ }
rank0.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9fbcef510669e7400b9111ba006d5589b17a1b60ab0f0d533acc297e843e386
3
+ size 16060549392