This model is a fine-tuned version of Qwen/Qwen3-8B on the penfever/glm-4.6-r2egym-32ep-32k, the penfever/GLM-4.6-nl2bash-verified-32eps-32k and the penfever/GLM-4.6-stackexchange-overflow-sandboxes-32eps-65k datasets.
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: 4e-05
train_batch_size: 1
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 2
total_train_batch_size: 16
total_eval_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
