Upload config.json with huggingface_hub
Browse files- config.json +34 -2
config.json
CHANGED
|
@@ -1,6 +1,38 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
"bos_token_id": 1,
|
| 4 |
"eos_token_id": 2,
|
| 5 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "/fs/archive/share/u20238046/falcon-rw-1b",
|
| 3 |
+
"alibi": true,
|
| 4 |
+
"apply_residual_connection_post_layernorm": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"FalconForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"auto_map": {
|
| 10 |
+
"AutoConfig": "configuration_falcon.FalconConfig",
|
| 11 |
+
"AutoModel": "modeling_falcon.FalconModel",
|
| 12 |
+
"AutoModelForCausalLM": "modeling_falcon.FalconForCausalLM",
|
| 13 |
+
"AutoModelForQuestionAnswering": "modeling_falcon.FalconForQuestionAnswering",
|
| 14 |
+
"AutoModelForSequenceClassification": "modeling_falcon.FalconForSequenceClassification",
|
| 15 |
+
"AutoModelForTokenClassification": "modeling_falcon.FalconForTokenClassification"
|
| 16 |
+
},
|
| 17 |
+
"bias": true,
|
| 18 |
"bos_token_id": 1,
|
| 19 |
"eos_token_id": 2,
|
| 20 |
+
"hidden_dropout": 0.0,
|
| 21 |
+
"hidden_size": 2048,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"layer_norm_epsilon": 1e-05,
|
| 24 |
+
"max_position_embeddings": 2048,
|
| 25 |
+
"model_type": "falcon",
|
| 26 |
+
"multi_query": false,
|
| 27 |
+
"new_decoder_architecture": false,
|
| 28 |
+
"num_attention_heads": 32,
|
| 29 |
+
"num_hidden_layers": 24,
|
| 30 |
+
"num_kv_heads": 32,
|
| 31 |
+
"parallel_attn": false,
|
| 32 |
+
"rope_scaling": null,
|
| 33 |
+
"rope_theta": 10000.0,
|
| 34 |
+
"torch_dtype": "bfloat16",
|
| 35 |
+
"transformers_version": "4.36.2",
|
| 36 |
+
"use_cache": true,
|
| 37 |
+
"vocab_size": 50304
|
| 38 |
}
|