This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-32B on the tttx/ttt-test-11-step1-child0 and the tttx/ttt-test-11-step1-master datasets. It achieves the following results on the evaluation set:

Loss: 0.0756

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: 4
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 32
total_eval_batch_size: 8
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_ratio: 0.1
num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss
0.1237	1.0	3	0.1191
0.0989	2.0	6	0.0898
0.0781	3.0	9	0.0775
0.0741	4.0	12	0.0756

Framework versions

PEFT 0.13.2
Transformers 4.47.0.dev0
Pytorch 2.4.0+cu121
Datasets 3.1.0
Tokenizers 0.20.3

tttx/models-ttt-test-11-step1

models-ttt-test-11-step1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

tttx/
models-ttt-test-11-step1