Upload model
Browse files- config.json +22 -0
- generation_config.json +4 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/data/sparrow/results/checkpoint-15000",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SparrowModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"dropout": 0.0,
|
| 8 |
+
"flash_attn": true,
|
| 9 |
+
"hidden_dim": 512,
|
| 10 |
+
"hidden_size": 512,
|
| 11 |
+
"intermediate_dim": 2048,
|
| 12 |
+
"max_seq_len": 512,
|
| 13 |
+
"mlp_bias": false,
|
| 14 |
+
"model_type": "sparrow",
|
| 15 |
+
"norm_eps": "1e-5",
|
| 16 |
+
"num_attention_heads": 16,
|
| 17 |
+
"num_hidden_layers": 8,
|
| 18 |
+
"num_key_value_heads": 16,
|
| 19 |
+
"torch_dtype": "float32",
|
| 20 |
+
"transformers_version": "4.48.2",
|
| 21 |
+
"vocab_size": 32000
|
| 22 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.48.2"
|
| 4 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:15cd79dd6a1cbfb8ae87135f083a03208d26a2d20311a9f71be12af4144ca3a0
|
| 3 |
+
size 269528160
|