Hugging Face's logo

reaperdoesntknow
/
Qemma-Q14B

Text Generation
Transformers
Safetensors
English
gemma3_text
trl
sft
gemma
qwen
Merge
disc
conversational
text-generation-inference
Model card Files
xet
Community
reaperdoesntknow commited on
Commit
b4ae528
·
verified ·
1 Parent(s): 0b8334b

Upload Gemma3ForCausalLM

Browse files
Files changed (3) hide show
  1. config.json +1 -4
  2. generation_config.json +1 -0
  3. model.safetensors +1 -1
config.json CHANGED
@@ -9,10 +9,7 @@
9
  "bos_token_id": 2,
10
  "cache_implementation": "sliding_window",
11
  "dtype": "float32",
12
- "eos_token_id": [
13
- 1,
14
- 106
15
- ],
16
  "final_logit_softcapping": null,
17
  "head_dim": 256,
18
  "hidden_activation": "gelu_pytorch_tanh",
 
9
  "bos_token_id": 2,
10
  "cache_implementation": "sliding_window",
11
  "dtype": "float32",
12
+ "eos_token_id": 1,
 
 
 
13
  "final_logit_softcapping": null,
14
  "head_dim": 256,
15
  "hidden_activation": "gelu_pytorch_tanh",
generation_config.json CHANGED
@@ -2,6 +2,7 @@
2
  "bos_token_id": 2,
3
  "do_sample": true,
4
  "eos_token_id": [
 
5
  1,
6
  106
7
  ],
 
2
  "bos_token_id": 2,
3
  "do_sample": true,
4
  "eos_token_id": [
5
+ 1,
6
  1,
7
  106
8
  ],
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:58225fb156b61b1b0b0e9e5c14198095bd8ffc406215db04e172612bedc317a0
3
  size 3999606000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:adce88e23254cc6b869aa5804b6e889ac513b022245411b2b816719a29823b45
3
  size 3999606000