Downloads · 30 days
8
11% of all-time downloads
open-llama-3b-opus-reasoning-sft-lora-2k
ping98k/open-llama-3b-opus-reasoning-sft-lora-2k
open-llama-3b-opus-reasoning-sft-lora-2k is a machine learning model from ping98k. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft.
LoRA adapter for openlm-research/openllama3bv2.
All-time downloads
75
Public
Repo size
1.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.4 GB · 100%
At a glance
- Library
- peft
- Access
- Public
- Created
- Apr 11, 2026
- Updated
- Apr 11, 2026
- SHA
- 349ed571
Base models
Model card
From the Hugging Face model README
open-llama-3b-opus-reasoning-sft-lora-2k
LoRA adapter for openlm-research/open_llama_3b_v2.
Training
- Base: open_llama_3b_v2 (4-bit NF4)
- Method: QLoRA (r=32, alpha=32)
- Data: Crownelius/Opus-4.6-Reasoning-3300x + Roman1111111/claude-opus-4.6-10000x
- Samples: 11,669 (filtered <= 2024 tokens)
- Epochs: 1
- Batch size: 8
- LR: 2e-4 (constant)
- Training: assistant-only (system/user turns masked)
- Special tokens: Qwen3.5 chat template (<|im_start|>, <|im_end|>, <think>, <tool_call>, etc.)
- modules_to_save: embed_tokens, lm_head
Format
<|im_start|>system
You are a helpful assistant.<|im_end|>
<|im_start|>user
{question}<|im_end|>
<|im_start|>assistant
<think>
{reasoning}
</think>
{answer}<|im_end|>
Merged model
See ping98k/open-llama-3b-opus-reasoning-sft-2k-4bit for the merged 4-bit model.
- Library
- peft
- Created
- Apr 11, 2026
- Updated
- Apr 11, 2026