metadata

library_name: transformers
license: other
base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
tags:
  - llama-factory
  - full
  - generated_from_trainer
model-index:
  - name: train_2025-01-23-00-42-56
    results: []

train_2025-01-23-00-42-56

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B on the smoltalk_chinese dataset. It achieves the following results on the evaluation set:

Loss: 1.9459

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: 0.0003
train_batch_size: 4
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 4
total_train_batch_size: 64
total_eval_batch_size: 16
optimizer: Use adamw_torch 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: 100
num_epochs: 2.0

Training results

Training Loss	Epoch	Step	Validation Loss
1.8719	1.7649	5000	1.9466

Framework versions

Transformers 4.46.1
Pytorch 2.5.1+cu124
Datasets 3.1.0
Tokenizers 0.20.3

MISC

_register_template(
    name="deepseekr1",
    default_system="You are a helpful and harmless assistant. You should think step-by-step. Output your thoughts in <think></think> tags.",
    format_prefix=EmptyFormatter(slots=[{"bos_token"}]),
    format_system=StringFormatter(slots=["{{content}}"]),
    format_user=StringFormatter(
        slots=[
            "<｜User｜>{{content}}<｜Assistant｜>"
        ]
    ),
    stop_words=["<｜end▁of▁sentence｜>"],
)