deepparag commited on
Commit
0149ae5
1 Parent(s): 0ee4724
Files changed (2) hide show
  1. config.json +7 -8
  2. pytorch_model.bin +2 -2
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "./output-small",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
@@ -8,14 +8,15 @@
8
  "bos_token_id": 50256,
9
  "embd_pdrop": 0.1,
10
  "eos_token_id": 50256,
 
11
  "initializer_range": 0.02,
12
  "layer_norm_epsilon": 1e-05,
13
  "model_type": "gpt2",
14
  "n_ctx": 1024,
15
- "n_embd": 1280,
16
- "n_head": 20,
17
  "n_inner": null,
18
- "n_layer": 36,
19
  "n_positions": 1024,
20
  "reorder_and_upcast_attn": false,
21
  "resid_pdrop": 0.1,
@@ -28,12 +29,10 @@
28
  "summary_use_proj": true,
29
  "task_specific_params": {
30
  "conversational": {
31
- "do_sample": true,
32
-
33
- "temperature": 0.8
34
  }
35
  },
36
- "torch_dtype": "float16",
37
  "transformers_version": "4.18.0",
38
  "use_cache": true,
39
  "vocab_size": 50257
 
1
  {
2
+ "_name_or_path": "bhaden94/LokiDiscordBot-medium",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
 
8
  "bos_token_id": 50256,
9
  "embd_pdrop": 0.1,
10
  "eos_token_id": 50256,
11
+ "gradient_checkpointing": false,
12
  "initializer_range": 0.02,
13
  "layer_norm_epsilon": 1e-05,
14
  "model_type": "gpt2",
15
  "n_ctx": 1024,
16
+ "n_embd": 1024,
17
+ "n_head": 16,
18
  "n_inner": null,
19
+ "n_layer": 24,
20
  "n_positions": 1024,
21
  "reorder_and_upcast_attn": false,
22
  "resid_pdrop": 0.1,
 
29
  "summary_use_proj": true,
30
  "task_specific_params": {
31
  "conversational": {
32
+ "max_length": 1000
 
 
33
  }
34
  },
35
+ "torch_dtype": "float32",
36
  "transformers_version": "4.18.0",
37
  "use_cache": true,
38
  "vocab_size": 50257
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:be625bdea4f890d79c5551532c4c9253b9cdab2687f562d8f030184b95fae997
3
- size 3134024393
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ece4de2ca6db4728c6a2942e05470d29ea75f98a3290f0f5ac19ed61f32f2e7c
3
+ size 1444562137