Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
gemma-3-12b-pt-full-pretrain-2048 – AI Model by ramzanniaz331 | AlphaNeural AI
You can deploy this model and start earning money today!
ramzanniaz331
/
gemma-3-12b-pt-full-pretrain-2048
like
0
transformers
safetensors
gemma3
image-text-to-text
llama-factory
full
generated_from_trainer
conversational
google/gemma-3-12b-pt
finetune
other
text-generation-inference
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
full_pretrain
This model is a fine-tuned version of
google/gemma-3-12b-pt
on the cpt_jazz_pretrain dataset. It achieves the following results on the evaluation set:
Loss: 1.9259
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
learning_rate: 5e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 32
total_train_batch_size: 256
total_eval_batch_size: 8
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.03
num_epochs: 1.0
Training results
Training Loss
Epoch
Step
Validation Loss
2.1161
0.1751
200
2.0828
2.0269
0.3502
400
2.0441
2.0119
0.5254
600
2.0156
1.9485
0.7005
800
1.9582
1.933
0.8756
1000
1.9292
Framework versions
Transformers 4.57.1
Pytorch 2.9.1+cu128
Datasets 4.0.0
Tokenizers 0.22.1
