You can deploy this model and start earning money today!

17Lab/
qwen3b-dpo-sft-s42

Name: qwen3b-dpo-sft-s42
Availability: InStock

17Lab/qwen3b-full-sft-s42

other

Views

No views yet

qwen3b_dpo_from_sft_s42

This model is a fine-tuned version of 17Lab/qwen3b-full-sft-s42 on the assimilation_dpo_v1 dataset. It achieves the following results on the evaluation set:

Loss: 0.2280
Rewards/chosen: -0.0231
Rewards/rejected: -1.9614
Rewards/accuracies: 0.9896
Rewards/margins: 1.9384
Logps/chosen: -10.0517
Logps/rejected: -41.9287
Logits/chosen: -2.7479
Logits/rejected: -2.7302

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-06
train_batch_size: 4
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 2
total_train_batch_size: 64
total_eval_batch_size: 16
optimizer: Use OptimizerNames.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: 0.1
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/chosen	Logps/rejected	Logits/chosen	Logits/rejected
0.2569	1.0	25	0.2280	-0.0231	-1.9614	0.9896	1.9384	-10.0517	-41.9287	-2.7479	-2.7302

Framework versions

PEFT 0.18.1
Transformers 5.6.0
Pytorch 2.7.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

17Lab/qwen3b-dpo-sft-s42

qwen3b_dpo_from_sft_s42

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

17Lab/
qwen3b-dpo-sft-s42