Upload MiniMoE
Browse files- config.json +23 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MiniMoE"
|
| 4 |
+
],
|
| 5 |
+
"auto_map": {
|
| 6 |
+
"AutoConfig": "mini_moe.MiniMoEConfig",
|
| 7 |
+
"AutoModelForCausalLM": "mini_moe.MiniMoE"
|
| 8 |
+
},
|
| 9 |
+
"dim": 1024,
|
| 10 |
+
"drop_rate": 0.0,
|
| 11 |
+
"dtype": "float32",
|
| 12 |
+
"model_type": "mini-moe",
|
| 13 |
+
"num_attention_kv_heads": 8,
|
| 14 |
+
"num_attention_q_heads": 16,
|
| 15 |
+
"num_expert": 8,
|
| 16 |
+
"num_layers": 12,
|
| 17 |
+
"qkv_bias": false,
|
| 18 |
+
"rope_base": 10000,
|
| 19 |
+
"top_k": 4,
|
| 20 |
+
"transformers_version": "4.56.2",
|
| 21 |
+
"use_aux_loss": false,
|
| 22 |
+
"vocab_size": 32000
|
| 23 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2624d4a4e2fa78f1df8796cadd1377e6f4bea255871f67f8d70c5ecd74b85c2
|
| 3 |
+
size 3506313467
|