sft__glm46-neulab-agenttuning-alfworld-sandboxes-maxeps-131k-glm46-swesmith-maxeps-131k-GLM-4-7
This model is a fine-tuned version of Qwen/Qwen3-8B on the /e/data1/datasets/playground/ot/hf_hub/datasets--penfever--glm46-neulab-agenttuning-alfworld-sandboxes-maxeps-131k/snapshots/fdb0d0afe08aa3c31c7605b40c18d5e48fdc206c_thinking_preprocessed, the /e/data1/datasets/playground/ot/hf_hub/datasets--penfever--glm46-swesmith-maxeps-131k/snapshots/4d4c2d4a9d21f73870ed31c7bc6028035b3b6ca7_thinking_preprocessed, the /e/data1/datasets/playground/ot/hf_hub/datasets--DCAgent2--GLM-4.7-r2egym_sandboxes-maxeps-131k/snapshots/167ff86e8203fa2412574480bf52623cb62320e8_thinking_preprocessed and the /e/data1/datasets/playground/ot/hf_hub/datasets--DCAgent2--glm46-swegym-tasks-maxeps-131k/snapshots/bc7a253d567261d84db295a138b8af86eac6ae4c_thinking_preprocessed 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: 48
- gradient_accumulation_steps: 2
- total_train_batch_size: 96
- total_eval_batch_size: 384
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 7.0
Training results
Framework versions
- Transformers 4.57.6
- Pytorch 2.9.1+cu130
- Datasets 4.7.0
- Tokenizers 0.22.2
- Downloads last month
- 39