Upload config.json with huggingface_hub
Browse files- config.json +32 -0
config.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"MolmoForCausalLM"
|
4 |
+
],
|
5 |
+
"attention_layer_norm": false,
|
6 |
+
"auto_map": {
|
7 |
+
"AutoConfig": "config_molmo.MolmoConfig",
|
8 |
+
"AutoModelForCausalLM": "modeling_molmo.MolmoForCausalLM"
|
9 |
+
},
|
10 |
+
"clip_qkv": null,
|
11 |
+
"embedding_size": 152064,
|
12 |
+
"hidden_size": 3584,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 37888,
|
15 |
+
"layer_norm_eps": 1e-06,
|
16 |
+
"layer_norm_type": "rms",
|
17 |
+
"max_position_embeddings": 4096,
|
18 |
+
"model_type": "molmo",
|
19 |
+
"norm_after": false,
|
20 |
+
"num_attention_heads": 28,
|
21 |
+
"num_hidden_layers": 28,
|
22 |
+
"num_key_value_heads": 4,
|
23 |
+
"qkv_bias": true,
|
24 |
+
"rope_theta": 1000000.0,
|
25 |
+
"tie_word_embeddings": false,
|
26 |
+
"torch_dtype": "float32",
|
27 |
+
"transformers_version": "4.43.3",
|
28 |
+
"use_cache": true,
|
29 |
+
"use_position_ids": true,
|
30 |
+
"vocab_size": 152064,
|
31 |
+
"weight_tying": false
|
32 |
+
}
|