Hugging Face's logo

lmassaron
/
gemma-3-270m-it-grpo-finsent-json

Text Generation
Transformers
Safetensors
gemma3_text
trl
grpo
unsloth
conversational
text-generation-inference
Model card Files
xet
Community
lmassaron commited on
Commit
df8a4c5
·
verified ·
1 Parent(s): 06a3ee6

Upload model trained with Unsloth

Browse files

Upload model trained with Unsloth 2x faster

Files changed (4) hide show
  1. README.md +1 -0
  2. config.json +20 -10
  3. generation_config.json +2 -1
  4. model.safetensors +2 -2
README.md CHANGED
@@ -3,6 +3,7 @@ library_name: transformers
3
  tags:
4
  - trl
5
  - grpo
 
6
  ---
7
 
8
  # Model Card for Model ID
 
3
  tags:
4
  - trl
5
  - grpo
6
+ - unsloth
7
  ---
8
 
9
  # Model Card for Model ID
config.json CHANGED
@@ -7,14 +7,14 @@
7
  "attention_dropout": 0.0,
8
  "attn_logit_softcapping": null,
9
  "bos_token_id": 2,
10
- "dtype": "bfloat16",
11
- "eos_token_id": 1,
12
  "final_logit_softcapping": null,
13
  "head_dim": 256,
14
  "hidden_activation": "gelu_pytorch_tanh",
15
- "hidden_size": 640,
16
  "initializer_range": 0.02,
17
- "intermediate_size": 2048,
18
  "layer_types": [
19
  "sliding_attention",
20
  "sliding_attention",
@@ -33,22 +33,32 @@
33
  "sliding_attention",
34
  "sliding_attention",
35
  "sliding_attention",
36
- "full_attention"
 
 
 
 
 
 
 
 
37
  ],
38
  "max_position_embeddings": 32768,
39
  "model_type": "gemma3_text",
40
  "num_attention_heads": 4,
41
- "num_hidden_layers": 18,
42
  "num_key_value_heads": 1,
43
  "pad_token_id": 0,
44
  "query_pre_attn_scalar": 256,
45
  "rms_norm_eps": 1e-06,
46
- "rope_local_base_freq": 10000.0,
47
  "rope_scaling": null,
48
- "rope_theta": 1000000.0,
49
  "sliding_window": 512,
50
- "transformers_version": "4.56.1",
51
- "use_bidirectional_attention": false,
 
 
52
  "use_cache": true,
53
  "vocab_size": 262144
54
  }
 
7
  "attention_dropout": 0.0,
8
  "attn_logit_softcapping": null,
9
  "bos_token_id": 2,
10
+ "cache_implementation": "hybrid",
11
+ "eos_token_id": 106,
12
  "final_logit_softcapping": null,
13
  "head_dim": 256,
14
  "hidden_activation": "gelu_pytorch_tanh",
15
+ "hidden_size": 1152,
16
  "initializer_range": 0.02,
17
+ "intermediate_size": 6912,
18
  "layer_types": [
19
  "sliding_attention",
20
  "sliding_attention",
 
33
  "sliding_attention",
34
  "sliding_attention",
35
  "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention"
45
  ],
46
  "max_position_embeddings": 32768,
47
  "model_type": "gemma3_text",
48
  "num_attention_heads": 4,
49
+ "num_hidden_layers": 26,
50
  "num_key_value_heads": 1,
51
  "pad_token_id": 0,
52
  "query_pre_attn_scalar": 256,
53
  "rms_norm_eps": 1e-06,
54
+ "rope_local_base_freq": 10000,
55
  "rope_scaling": null,
56
+ "rope_theta": 1000000,
57
  "sliding_window": 512,
58
+ "torch_dtype": "bfloat16",
59
+ "transformers_version": "4.55.4",
60
+ "unsloth_fixed": true,
61
+ "unsloth_version": "2025.9.7",
62
  "use_cache": true,
63
  "vocab_size": 262144
64
  }
generation_config.json CHANGED
@@ -6,8 +6,9 @@
6
  1,
7
  106
8
  ],
 
9
  "pad_token_id": 0,
10
  "top_k": 64,
11
  "top_p": 0.95,
12
- "transformers_version": "4.56.1"
13
  }
 
6
  1,
7
  106
8
  ],
9
+ "max_length": 32768,
10
  "pad_token_id": 0,
11
  "top_k": 64,
12
  "top_p": 0.95,
13
+ "transformers_version": "4.55.4"
14
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:412ca71266d94d6122ff4527cb50a1a536af21dbb23ed997e35038bd2cffb4eb
3
- size 536223056
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39a650d1cd96b76d14d905dbe78aa52a8123c37e78fdf9a9e8a34772c5c30efd
3
+ size 2000078368