LLM EXPLORER 59,358 MODELS INDEXED

Qwen 0.5 RDPO 5e 7 3ep 0vpo Const 0.3 by JayHyeon

By JayHyeon · 6 downloads
HF Repo ↗ Copy URL

Qwen 0.5 RDPO 5e 7 3ep 0vpo Const 0.3 is an open-source language model by JayHyeon. Features: 0.5b LLM, VRAM: 1.3GB, Context: 32K, LLM Explorer Score: 0.19.

  Arxiv:2305.18290 Base model:finetune:jayhyeon/q... Base model:jayhyeon/qwen2.5-0....   Conversational Dataset:trl-lib/ultrafeedback ...   Dpo   Endpoints compatible   Generated from trainer   Qwen2   Region:us   Safetensors   Trl
Model Card on HF 🤗: https://huggingface.co/JayHyeon/Qwen_0.5-rDPO_5e-7-3ep_0vpo_const_0.3 

Qwen 0.5 RDPO 5e 7 3ep 0vpo Const 0.3 Parameters and Internals

LLM NameQwen 0.5 RDPO 5e 7 3ep 0vpo Const 0.3
Repository 🤗https://huggingface.co/JayHyeon/Qwen_0.5-rDPO_5e-7-3ep_0vpo_const_0.3 
Model NameQwen_0.5-rDPO_5e-7-3ep_0vpo_const
Base Model(s)  JayHyeon/Qwen2.5-0.5B-SFT-2e-5-2ep   JayHyeon/Qwen2.5-0.5B-SFT-2e-5-2ep
Model Size0.5b
Required VRAM1.3 GB
Updated2026-06-08
MaintainerJayHyeon
Model Typeqwen2
Model Files  1.3 GB   0.0 GB
Model ArchitectureQwen2ForCausalLM
Context Length32768
Model Max Length32768
Transformers Version4.47.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Qwen 0.5 RDPO 5e 7 3ep 0vpo Const 0.3

Best Alternatives
Context / RAM
Downloads
Likes
Reader Lm 0.5B250K / 1 GB188150
Qwen2 0.5B128K / 1 GB1216780169
Qwen2 0.5B Finetune Exp 1128K / 1 GB150
...0.5B FT EnhancedMUSREnsembleV3128K / 1 GB321
....5B FT V4 MUSR ENSEMBLE Mathis128K / 2 GB301
...0.5B FT MUSR ENSEMBLE V2Mathis128K / 1 GB250
...PRYMMAL 0.5B FT V4 MUSR Mathis128K / 1 GB621
Qwen2 0.5B Stf Imdb128K / 1 GB70
ECE PRYMMAL 0.5B FT V3 MUSR128K / 2 GB170
ECE PRYMMAL 0.5B FT V4 MUSR128K / 2 GB100
Note: green Score (e.g. "73.2") means that the model is better than JayHyeon/Qwen_0.5-rDPO_5e-7-3ep_0vpo_const_0.3.