Initial upload of custom GPT model
Browse files- config.json +23 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vocab_size": 50257,
|
| 3 |
+
"embedding_dimension": 512,
|
| 4 |
+
"num_heads": 8,
|
| 5 |
+
"context_length": 256,
|
| 6 |
+
"token_dropout": 0.03,
|
| 7 |
+
"attn_dropout": 0.2,
|
| 8 |
+
"ffn_dropout": 0.2,
|
| 9 |
+
"qkv_bias": false,
|
| 10 |
+
"num_layers": 8,
|
| 11 |
+
"ff_hidden_dim": 1024,
|
| 12 |
+
"rms_eps": 1e-06,
|
| 13 |
+
"rms_bias": true,
|
| 14 |
+
"theta_base": 10000.0,
|
| 15 |
+
"num_kv_groups": 4,
|
| 16 |
+
"num_experts": 4,
|
| 17 |
+
"num_active_experts": 2,
|
| 18 |
+
"moe_noise": true,
|
| 19 |
+
"architectures": [
|
| 20 |
+
"GQAGPT2"
|
| 21 |
+
],
|
| 22 |
+
"model_type": "customGPT_pretrain"
|
| 23 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ff9ba28caefa429b99c98be2dafc5d2b7b0fd38d06102b2887d78e3ee918bf4
|
| 3 |
+
size 195393712
|