jethac commited on
Commit
73ce86c
·
verified ·
1 Parent(s): 9602849

Training in progress, epoch 5

Browse files
README.md CHANGED
@@ -1,17 +1,17 @@
1
  ---
2
- base_model: google/gemma-3-270m-it
3
  library_name: transformers
4
  model_name: MyGemmaNPC
5
  tags:
6
  - generated_from_trainer
7
- - trl
8
  - sft
 
9
  licence: license
10
  ---
11
 
12
  # Model Card for MyGemmaNPC
13
 
14
- This model is a fine-tuned version of [google/gemma-3-270m-it](https://huggingface.co/google/gemma-3-270m-it).
15
  It has been trained using [TRL](https://github.com/huggingface/trl).
16
 
17
  ## Quick start
 
1
  ---
2
+ base_model: google/gemma-3-1b-it
3
  library_name: transformers
4
  model_name: MyGemmaNPC
5
  tags:
6
  - generated_from_trainer
 
7
  - sft
8
+ - trl
9
  licence: license
10
  ---
11
 
12
  # Model Card for MyGemmaNPC
13
 
14
+ This model is a fine-tuned version of [google/gemma-3-1b-it](https://huggingface.co/google/gemma-3-1b-it).
15
  It has been trained using [TRL](https://github.com/huggingface/trl).
16
 
17
  ## Quick start
config.json CHANGED
@@ -7,13 +7,17 @@
7
  "attention_dropout": 0.0,
8
  "attn_logit_softcapping": null,
9
  "bos_token_id": 2,
10
- "eos_token_id": 1,
 
 
 
 
11
  "final_logit_softcapping": null,
12
  "head_dim": 256,
13
  "hidden_activation": "gelu_pytorch_tanh",
14
- "hidden_size": 640,
15
  "initializer_range": 0.02,
16
- "intermediate_size": 2048,
17
  "layer_types": [
18
  "sliding_attention",
19
  "sliding_attention",
@@ -32,23 +36,30 @@
32
  "sliding_attention",
33
  "sliding_attention",
34
  "sliding_attention",
35
- "full_attention"
 
 
 
 
 
 
 
 
36
  ],
37
  "max_position_embeddings": 32768,
38
  "model_type": "gemma3_text",
39
  "num_attention_heads": 4,
40
- "num_hidden_layers": 18,
41
  "num_key_value_heads": 1,
42
  "pad_token_id": 0,
43
  "query_pre_attn_scalar": 256,
44
  "rms_norm_eps": 1e-06,
45
- "rope_local_base_freq": 10000.0,
46
  "rope_scaling": null,
47
- "rope_theta": 1000000.0,
48
  "sliding_window": 512,
49
  "torch_dtype": "bfloat16",
50
  "transformers_version": "4.55.2",
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": [
12
+ 1,
13
+ 106
14
+ ],
15
  "final_logit_softcapping": null,
16
  "head_dim": 256,
17
  "hidden_activation": "gelu_pytorch_tanh",
18
+ "hidden_size": 1152,
19
  "initializer_range": 0.02,
20
+ "intermediate_size": 6912,
21
  "layer_types": [
22
  "sliding_attention",
23
  "sliding_attention",
 
36
  "sliding_attention",
37
  "sliding_attention",
38
  "sliding_attention",
39
+ "full_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "full_attention",
46
+ "sliding_attention",
47
+ "sliding_attention"
48
  ],
49
  "max_position_embeddings": 32768,
50
  "model_type": "gemma3_text",
51
  "num_attention_heads": 4,
52
+ "num_hidden_layers": 26,
53
  "num_key_value_heads": 1,
54
  "pad_token_id": 0,
55
  "query_pre_attn_scalar": 256,
56
  "rms_norm_eps": 1e-06,
57
+ "rope_local_base_freq": 10000,
58
  "rope_scaling": null,
59
+ "rope_theta": 1000000,
60
  "sliding_window": 512,
61
  "torch_dtype": "bfloat16",
62
  "transformers_version": "4.55.2",
 
63
  "use_cache": true,
64
  "vocab_size": 262144
65
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6b10557428c98c0639986a03e71013efb7cdb38b40983d9048aa485cef784e3c
3
- size 536223056
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c054f84899381e2473377c93eaafe54b2add62fdb31915f7afb26c6cbb8f6069
3
+ size 1999811208
runs/Aug22_07-30-47_4891721524f3/events.out.tfevents.1755847851.4891721524f3.5316.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea856d1a0578996ed7d43f91f79f275d94610c2841ac679c915cbebcecce53b8
3
+ size 16777
tokenizer_config.json CHANGED
@@ -51336,7 +51336,7 @@
51336
  "image_token": "<image_soft_token>",
51337
  "model_max_length": 1000000000000000019884624838656,
51338
  "pad_token": "<pad>",
51339
- "padding_side": "left",
51340
  "sp_model_kwargs": null,
51341
  "spaces_between_special_tokens": false,
51342
  "tokenizer_class": "GemmaTokenizer",
 
51336
  "image_token": "<image_soft_token>",
51337
  "model_max_length": 1000000000000000019884624838656,
51338
  "pad_token": "<pad>",
51339
+ "processor_class": "Gemma3Processor",
51340
  "sp_model_kwargs": null,
51341
  "spaces_between_special_tokens": false,
51342
  "tokenizer_class": "GemmaTokenizer",
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:903c5427a8abd862e3fa06a092f52aca9ee07f4732f13cae585f56a5670841c7
3
  size 6225
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da2715a022e092521fe8785518c0d3b8af3c8be6714032feb15c645a2893f7ea
3
  size 6225