Commit
•
3df62ce
1
Parent(s):
95dda18
Upload GPTNeoXForCausalLM
Browse files- config.json +7 -9
- pytorch_model.bin +3 -0
config.json
CHANGED
@@ -1,21 +1,19 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"architectures": [
|
4 |
"GPTNeoXForCausalLM"
|
5 |
],
|
6 |
-
"attention_probs_dropout_prob": 0,
|
7 |
"bos_token_id": 0,
|
8 |
"eos_token_id": 0,
|
9 |
-
"hidden_act": "
|
10 |
-
"
|
11 |
-
"hidden_size": 6144,
|
12 |
"initializer_range": 0.02,
|
13 |
-
"intermediate_size":
|
14 |
"layer_norm_eps": 1e-05,
|
15 |
"max_position_embeddings": 2048,
|
16 |
"model_type": "gpt_neox",
|
17 |
-
"num_attention_heads":
|
18 |
-
"num_hidden_layers":
|
19 |
"rotary_emb_base": 10000,
|
20 |
"rotary_pct": 0.25,
|
21 |
"tie_word_embeddings": false,
|
@@ -23,5 +21,5 @@
|
|
23 |
"transformers_version": "4.28.1",
|
24 |
"use_cache": true,
|
25 |
"use_parallel_residual": true,
|
26 |
-
"vocab_size":
|
27 |
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "maomao0301/pythia-1b-imdb-adapter-merged",
|
3 |
"architectures": [
|
4 |
"GPTNeoXForCausalLM"
|
5 |
],
|
|
|
6 |
"bos_token_id": 0,
|
7 |
"eos_token_id": 0,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_size": 2048,
|
|
|
10 |
"initializer_range": 0.02,
|
11 |
+
"intermediate_size": 8192,
|
12 |
"layer_norm_eps": 1e-05,
|
13 |
"max_position_embeddings": 2048,
|
14 |
"model_type": "gpt_neox",
|
15 |
+
"num_attention_heads": 8,
|
16 |
+
"num_hidden_layers": 16,
|
17 |
"rotary_emb_base": 10000,
|
18 |
"rotary_pct": 0.25,
|
19 |
"tie_word_embeddings": false,
|
|
|
21 |
"transformers_version": "4.28.1",
|
22 |
"use_cache": true,
|
23 |
"use_parallel_residual": true,
|
24 |
+
"vocab_size": 50304
|
25 |
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6a15b35016a52c0a50155bb8ed410ef4efd08e53f78b2dd8afbb22f5fa8e856b
|
3 |
+
size 2090752989
|