Upload model
Browse files- config.json +36 -0
- generation_config.json +4 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"AutoModelForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"attn_pdrop": 0.0,
|
| 8 |
+
"embd_pdrop": 0.0,
|
| 9 |
+
"flash_attn": false,
|
| 10 |
+
"flash_rotary": false,
|
| 11 |
+
"fused_dense": false,
|
| 12 |
+
"hidden_act": "gelu_new",
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"layer_norm_epsilon": 1e-05,
|
| 17 |
+
"model_type": "phi",
|
| 18 |
+
"n_embd": 2048,
|
| 19 |
+
"n_head": 32,
|
| 20 |
+
"n_head_kv": null,
|
| 21 |
+
"n_inner": null,
|
| 22 |
+
"n_layer": 24,
|
| 23 |
+
"n_positions": 2048,
|
| 24 |
+
"num_key_value_heads": null,
|
| 25 |
+
"partial_rotary_factor": 0.5,
|
| 26 |
+
"qk_layernorm": false,
|
| 27 |
+
"resid_pdrop": 0.0,
|
| 28 |
+
"rope_scaling": null,
|
| 29 |
+
"rope_theta": 10000.0,
|
| 30 |
+
"rotary_dim": 32,
|
| 31 |
+
"tie_word_embeddings": false,
|
| 32 |
+
"torch_dtype": "bfloat16",
|
| 33 |
+
"transformers_version": "4.50.0",
|
| 34 |
+
"use_cache": true,
|
| 35 |
+
"vocab_size": 51200
|
| 36 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.50.0"
|
| 4 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09d0db49cbbc735576cf0b7ade2fabfa2334e55c1f15d913851259846718f37e
|
| 3 |
+
size 2836567400
|