LLM EXPLORER 67,390 MODELS INDEXED

Phi 2 Instruct by venkycs

By venkycs · 124 downloads

Phi 2 Instruct is an open-source language model by venkycs. Features: 2.8b LLM, VRAM: 11.1GB, License: mit, Instruction-Based, LLM Explorer Score: 0.1.

  Adapter   Adapter-transformers Base model:adapter:microsoft/p...   Base model:microsoft/phi-2   Custom code Dataset:huggingfaceh4/ultracha...   En   Finetuned   Generated from trainer   Instruct   Lora   Phi-msft   Region:us   Safetensors   Sharded   Tensorboard   Tensorflow
Model Card on HF πŸ€—: https://huggingface.co/venkycs/phi-2-instruct 

Phi 2 Instruct Parameters and Internals

Model Type 
text-generation
Additional Notes 
This model is a fine-tuned version of microsoft/phi-2 on the filtered ultrachat200k dataset using the SFT technique. More information about model architecture and specific modifications made during fine-tuning is needed.
Supported Languages 
en (default)
Training Details 
Data Sources:
HuggingFaceH4/ultrachat_200k
Methodology:
SFT technique
LLM NamePhi 2 Instruct
Repository πŸ€—https://huggingface.co/venkycs/phi-2-instruct 
Base Model(s)  Phi 2   microsoft/phi-2
Model Size2.8b
Required VRAM11.1 GB
Updated2026-06-02
Maintainervenkycs
Instruction-BasedYes
Model Files  0.1 GB   4.9 GB: 1-of-3   4.9 GB: 2-of-3   1.3 GB: 3-of-3   0.0 GB
Supported Languagesen
Model ArchitectureAdapter
Licensemit
Model Max Length2048
Is Biasednone
Tokenizer ClassCodeGenTokenizer
PEFT TypeLORA
LoRA ModelYes
PEFT Target Modulesout_proj|fc1|Wqkv|fc2
LoRA Alpha16
LoRA Dropout0.05
R Param16