AndreyRzhaksinskiy
commited on
Commit
•
ac36f78
1
Parent(s):
22ca110
Upload Starcoder2ForCausalLM
Browse files- config.json +3 -2
- generation_config.json +1 -1
- model-00001-of-00003.safetensors +1 -1
- model-00002-of-00003.safetensors +1 -1
- model-00003-of-00003.safetensors +1 -1
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"activation_function": "gelu",
|
4 |
"architectures": [
|
5 |
"Starcoder2ForCausalLM"
|
@@ -23,12 +23,13 @@
|
|
23 |
"num_hidden_layers": 32,
|
24 |
"num_key_value_heads": 4,
|
25 |
"residual_dropout": 0.1,
|
|
|
26 |
"rope_theta": 1000000,
|
27 |
"scale_attention_softmax_in_fp32": true,
|
28 |
"scale_attn_weights": true,
|
29 |
"sliding_window": 4096,
|
30 |
"torch_dtype": "float16",
|
31 |
-
"transformers_version": "4.
|
32 |
"use_bias": true,
|
33 |
"use_cache": true,
|
34 |
"vocab_size": 49152
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "bigcode/starcoder2-7b",
|
3 |
"activation_function": "gelu",
|
4 |
"architectures": [
|
5 |
"Starcoder2ForCausalLM"
|
|
|
23 |
"num_hidden_layers": 32,
|
24 |
"num_key_value_heads": 4,
|
25 |
"residual_dropout": 0.1,
|
26 |
+
"rope_scaling": null,
|
27 |
"rope_theta": 1000000,
|
28 |
"scale_attention_softmax_in_fp32": true,
|
29 |
"scale_attn_weights": true,
|
30 |
"sliding_window": 4096,
|
31 |
"torch_dtype": "float16",
|
32 |
+
"transformers_version": "4.45.2",
|
33 |
"use_bias": true,
|
34 |
"use_cache": true,
|
35 |
"vocab_size": 49152
|
generation_config.json
CHANGED
@@ -2,5 +2,5 @@
|
|
2 |
"_from_model_config": true,
|
3 |
"bos_token_id": 0,
|
4 |
"eos_token_id": 0,
|
5 |
-
"transformers_version": "4.
|
6 |
}
|
|
|
2 |
"_from_model_config": true,
|
3 |
"bos_token_id": 0,
|
4 |
"eos_token_id": 0,
|
5 |
+
"transformers_version": "4.45.2"
|
6 |
}
|
model-00001-of-00003.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4889534784
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8c1e5cd5426894cba27ef5eb532e3e107ec0956ebc746d14e425f6617a5b988e
|
3 |
size 4889534784
|
model-00002-of-00003.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4946278808
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:19aa42ad9b65fa2b15e3c2e08bcfcbeee38891c3869de4f22fe3f78867752f11
|
3 |
size 4946278808
|
model-00003-of-00003.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4512091408
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a1218418e9b986867349386521d7cf000b7343937c7c27e7e2cb8b2e00d0d085
|
3 |
size 4512091408
|