Commit
•
2e94147
1
Parent(s):
c7fd6db
Upload FalconForCausalLM
Browse files- config.json +29 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/workspace/pretraining/local-models/training/2023-12-31_20-30-43",
|
| 3 |
+
"alibi": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"FalconForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bias": false,
|
| 9 |
+
"bos_token_id": 11,
|
| 10 |
+
"eos_token_id": 11,
|
| 11 |
+
"hidden_dropout": 0.0,
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"layer_norm_epsilon": 1e-05,
|
| 15 |
+
"max_position_embeddings": 1024,
|
| 16 |
+
"model_type": "falcon",
|
| 17 |
+
"multi_query": true,
|
| 18 |
+
"new_decoder_architecture": false,
|
| 19 |
+
"num_attention_heads": 64,
|
| 20 |
+
"num_hidden_layers": 4,
|
| 21 |
+
"num_kv_heads": 64,
|
| 22 |
+
"parallel_attn": true,
|
| 23 |
+
"rope_scaling": null,
|
| 24 |
+
"rope_theta": 10000.0,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.34.1",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"vocab_size": 65024
|
| 29 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 11,
|
| 4 |
+
"eos_token_id": 11,
|
| 5 |
+
"transformers_version": "4.34.1"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:486b986ae690aee5d4bfbe8ea33a216e98f3cc9fa3254f409aca1744fee23f3f
|
| 3 |
+
size 434678784
|