Downloads · 30 days
0
Qwen3.5-2B-torch-dist
lcj111wl/Qwen3.5-2B-torch-dist
Qwen3.5-2B-torch-dist is a text generation model from lcj111wl. Use it when you need the model to write or continue text. It is set up for megatron-lm. The card lists the license as other.
This repository contains a Megatron-LM torchdist checkpoint converted from a local Hugging Face safetensors checkpoint for use with slime/Megatron SFT training.
Access
Public
Updated Jun 12, 2026
Repo size
3.8 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.distcp3.8 GB · 100%
At a glance
- Task
- Text Generation
- Library
- megatron-lm
- License
- other
- Access
- Public
- Created
- Jun 12, 2026
- Updated
- Jun 12, 2026
- SHA
- f3687b04
Try a prompt
Base models
Model card
From the Hugging Face model README
Qwen3.5-2B Megatron torch_dist checkpoint
This repository contains a Megatron-LM torch_dist checkpoint converted from a local Hugging Face safetensors checkpoint for use with slime/Megatron SFT training.
Contents
latest_checkpointed_iteration.txt->releaserelease/Megatron distributed checkpoint directoryrun_convert_qwen35_2b.shconversion command used onmachine2qwen3.5-2B.shslime/Megatron model args used for conversion
Conversion Notes
- Source local archive:
/Users/lcj/Downloads/Qwen3.5-2B.zip - Remote conversion workdir:
/home/ubuntu/workspace/lcj/qwen35_2b_torch_dist - Megatron-LM commit:
1dcf0dafa884ad52ffb243625717a3471643e087 - Torch dtype: bf16
- Tensor parallel size: 1
- Pipeline parallel size: 1
- Tied embeddings: enabled (
untie_embeddings_and_output_weights=False)
- Task
- Text Generation
- Library
- megatron-lm
- License
- other
- Created
- Jun 12, 2026
- Updated
- Jun 12, 2026