Faro Qwen 1.8b model
Browse files- config.json +31 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/cephfs/panwenbo/work/repos/fusang_8x7b/outputs/fi-chatml-1.8/merged",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Qwen2Model"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"eos_token_id": 151645,
|
| 8 |
+
"hidden_act": "silu",
|
| 9 |
+
"hidden_size": 2048,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 5504,
|
| 12 |
+
"max_position_embeddings": 32768,
|
| 13 |
+
"max_window_layers": 21,
|
| 14 |
+
"model_type": "qwen2",
|
| 15 |
+
"num_attention_heads": 16,
|
| 16 |
+
"num_hidden_layers": 24,
|
| 17 |
+
"num_key_value_heads": 16,
|
| 18 |
+
"rms_norm_eps": 1e-06,
|
| 19 |
+
"rope_scaling": {
|
| 20 |
+
"factor": 4.0,
|
| 21 |
+
"type": "dynamic"
|
| 22 |
+
},
|
| 23 |
+
"rope_theta": 1000000.0,
|
| 24 |
+
"sliding_window": 32768,
|
| 25 |
+
"tie_word_embeddings": false,
|
| 26 |
+
"torch_dtype": "bfloat16",
|
| 27 |
+
"transformers_version": "4.39.3",
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"use_sliding_window": false,
|
| 30 |
+
"vocab_size": 151936
|
| 31 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b7e8c6e3eeafbfe5f3e188121445c2819a44c20b187c92674e6d78d469d1d13
|
| 3 |
+
size 3051358920
|