You can deploy this model and start earning money today!

xummer/
deepseek-r1-8b-belebele-lora-tam-taml

Name: deepseek-r1-8b-belebele-lora-tam-taml
Availability: InStock

deepseek-ai/DeepSeek-R1-Distill-Llama-8B

other

Views

No views yet

belebele_tam_Taml

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Llama-8B on the belebele_tam_Taml_train dataset. It achieves the following results on the evaluation set:

Loss: 0.5502
Accuracy: 0.8912
Mcq Accuracy: 0.3

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: 0.0002
train_batch_size: 16
eval_batch_size: 16
seed: 42
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_steps: 0.1
num_epochs: 5.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Mcq Accuracy
0.3133	1.0	40	0.3762	0.8871	0.1667
0.1893	2.0	80	0.3429	0.9012	0.3778
0.1187	3.0	120	0.4063	0.8892	0.3111
0.0432	4.0	160	0.4828	0.8920	0.3111
0.0156	5.0	200	0.5502	0.8912	0.3

Framework versions

PEFT 0.18.1
Transformers 5.2.0
Pytorch 2.10.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

xummer/deepseek-r1-8b-belebele-lora-tam-taml

belebele_tam_Taml

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

xummer/
deepseek-r1-8b-belebele-lora-tam-taml