Upload folder using huggingface_hub
Browse files- config.json +40 -0
- rank0.safetensors +3 -0
config.json
ADDED
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"producer": {
|
3 |
+
"name": "modelopt",
|
4 |
+
"version": "0.13.0"
|
5 |
+
},
|
6 |
+
"architecture": "LlamaForCausalLM",
|
7 |
+
"dtype": "float16",
|
8 |
+
"logits_dtype": "float16",
|
9 |
+
"num_hidden_layers": 32,
|
10 |
+
"num_attention_heads": 32,
|
11 |
+
"num_key_value_heads": 32,
|
12 |
+
"hidden_size": 4096,
|
13 |
+
"norm_epsilon": 1e-05,
|
14 |
+
"vocab_size": 32000,
|
15 |
+
"max_position_embeddings": 4096,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"use_parallel_embedding": true,
|
18 |
+
"embedding_sharding_dim": 0,
|
19 |
+
"quantization": {
|
20 |
+
"quant_algo": "W8A8_SQ_PER_CHANNEL",
|
21 |
+
"kv_cache_quant_algo": "FP8"
|
22 |
+
},
|
23 |
+
"mapping": {
|
24 |
+
"world_size": 1,
|
25 |
+
"tp_size": 1,
|
26 |
+
"pp_size": 1
|
27 |
+
},
|
28 |
+
"head_size": 128,
|
29 |
+
"intermediate_size": 11008,
|
30 |
+
"position_embedding_type": "rope_gpt_neox",
|
31 |
+
"share_embedding_table": false,
|
32 |
+
"residual_mlp": false,
|
33 |
+
"bias": false,
|
34 |
+
"rotary_pct": 1.0,
|
35 |
+
"rank": 0,
|
36 |
+
"decoder": "llama",
|
37 |
+
"rmsnorm": true,
|
38 |
+
"lm_head_bias": false,
|
39 |
+
"rotary_base": 10000.0
|
40 |
+
}
|
rank0.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:213e3e7c844c796316186826f227f8a67d039e218652fbc0d6e4eeb30dbf7687
|
3 |
+
size 13476861480
|