LLM EXPLORER 68,292 MODELS INDEXED

Llama3.1 8B Instruct Convfinqa RFT Sc K10 Lora R16a32 Lr1e 4 Ep3 Ratio0.2 by Jongbin-kr

By Jongbin-kr · 15 downloads

Llama3.1 8B Instruct Convfinqa RFT Sc K10 Lora R16a32 Lr1e 4 Ep3 Ratio0.2 is an open-source language model by Jongbin-kr. Features: 8b LLM, VRAM: 0.2GB, License: llama3.1, Instruction-Based.

  Adapter Base model:adapter:meta-llama/... Base model:meta-llama/llama-3....   Conversational   Convfinqa   Finetuned   Instruct   Jev   Lora   Peft   Region:us   Rejection-sampling-fine-tuning   Rft   Safetensors   Self-training

Llama3.1 8B Instruct Convfinqa RFT Sc K10 Lora R16a32 Lr1e 4 Ep3 Ratio0.2 Parameters and Internals

LLM NameLlama3.1 8B Instruct Convfinqa RFT Sc K10 Lora R16a32 Lr1e 4 Ep3 Ratio0.2
Repository πŸ€—https://huggingface.co/Jongbin-kr/llama3.1-8b-instruct_convfinqa_RFT-sc_k10_lora-r16a32_lr1e-4_ep3_ratio0.2 
Base Model(s)  meta-llama/Llama-3.1-8B-Instruct   meta-llama/Llama-3.1-8B-Instruct
Model Size8b
Required VRAM0.2 GB
Updated2026-11-22
MaintainerJongbin-kr
Instruction-BasedYes
Model Files  0.2 GB   0.0 GB
Model ArchitectureAdapter
Licensellama3.1
Model Max Length131072
Is Biasednone
Tokenizer ClassTokenizersBackend
Padding Token<|finetune_right_pad_id|>
PEFT TypeLORA
LoRA ModelYes
PEFT Target Modulesup_proj|o_proj|q_proj|down_proj|v_proj|k_proj|gate_proj
LoRA Alpha32
LoRA Dropout0.05
R Param16

Best Alternatives to Llama3.1 8B Instruct Convfinqa RFT Sc K10 Lora R16a32 Lr1e 4 Ep3 Ratio0.2

Best Alternatives
Context / RAM
Downloads
Likes
... 3 8B Instruct Bvr Finetune V38K / 16.1 GB100
Visual Jev 8B Answer Sft0K / 0.2 GB441
K Oss 8 Mini Instruct Beta0K / 0.2 GB501
...Max H3 Prompt Rewriter LoRA 8B0K / 2.8 GB13726
Velora V10K / 0 GB261
Llama3 8B Persian Lora0K / 0.2 GB221
Genlocalnews Llama3.1 8B Lora0K / 0.2 GB211
....1 8B Reponx Network Cloud Ops0K / 0.2 GB142
...27 Llama31 8B Sgd New Rank 1280K / 1.3 GB201
VitaGridProtocol0K /  GB151
Note: green Score (e.g. "73.2") means that the model is better than Jongbin-kr/llama3.1-8b-instruct_convfinqa_RFT-sc_k10_lora-r16a32_lr1e-4_ep3_ratio0.2.