Upload train_olmo/config.json with huggingface_hub
Browse files- train_olmo/config.json +32 -0
train_olmo/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MolmoForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_layer_norm": true,
|
| 6 |
+
"auto_map": {
|
| 7 |
+
"AutoConfig": "config_molmo.MolmoConfig",
|
| 8 |
+
"AutoModelForCausalLM": "allenai/Molmo-7B-O-0924--modeling_molmo.MolmoForCausalLM"
|
| 9 |
+
},
|
| 10 |
+
"clip_qkv": null,
|
| 11 |
+
"embedding_size": 100352,
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 22016,
|
| 15 |
+
"layer_norm_eps": 1e-06,
|
| 16 |
+
"layer_norm_type": "rms",
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"model_type": "molmo",
|
| 19 |
+
"norm_after": true,
|
| 20 |
+
"num_attention_heads": 32,
|
| 21 |
+
"num_hidden_layers": 32,
|
| 22 |
+
"num_key_value_heads": null,
|
| 23 |
+
"qkv_bias": false,
|
| 24 |
+
"rope_theta": 500000.0,
|
| 25 |
+
"tie_word_embeddings": false,
|
| 26 |
+
"torch_dtype": "float32",
|
| 27 |
+
"transformers_version": "4.50.3",
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"use_position_ids": true,
|
| 30 |
+
"vocab_size": 100278,
|
| 31 |
+
"weight_tying": false
|
| 32 |
+
}
|