Dejiao Z
commited on
Commit
·
55bfc22
1
Parent(s):
b63560b
init push
Browse files- config.json +31 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/mnt/efs/people/dejiaoz/experiments/next_models/codesage_v2_pub/cs-small-clv7-fv1-3e05-4k18k-mlmdobf-para-homo-t003-mrl/combined_bf16_unwrapped.pt",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"CodeSageModel"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout_prob": 0.1,
|
| 8 |
+
"auto_map": {
|
| 9 |
+
"AutoConfig": "config_codesage.CodeSageConfig",
|
| 10 |
+
"AutoModel": "modeling_codesage.CodeSageModel",
|
| 11 |
+
"AutoModelForMaskedLM": "modeling_codesage.CodeSageForMaskedLM",
|
| 12 |
+
"AutoModelForSequenceClassification": "modeling_codesage.CodeSageForSequenceClassification"
|
| 13 |
+
},
|
| 14 |
+
"bos_token_id": 100257,
|
| 15 |
+
"embedding_dropout_prob": 0.1,
|
| 16 |
+
"eos_token_id": 100257,
|
| 17 |
+
"hidden_size": 1024,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 4096,
|
| 20 |
+
"layer_norm_epsilon": 1e-05,
|
| 21 |
+
"max_position_embeddings": 2048,
|
| 22 |
+
"model_type": "codesage",
|
| 23 |
+
"num_attention_heads": 8,
|
| 24 |
+
"num_hidden_layers": 6,
|
| 25 |
+
"pad_token_id": 100317,
|
| 26 |
+
"position_embedding_type": "absolute",
|
| 27 |
+
"residual_dropout_prob": 0.1,
|
| 28 |
+
"torch_dtype": "bfloat16",
|
| 29 |
+
"transformers_version": "4.26.1",
|
| 30 |
+
"vocab_size": 49154
|
| 31 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6ed7fe6ee8a40b7830482c91421733f4c66c6e128adfc87dd9cee1a41936c14
|
| 3 |
+
size 256041857
|