ONNX Model Zoo

martinkorelic/Qwen2-0.5B-MiniRecommendation-mobile-trained

By martinkorelic · ↓ 0 · ♥ 0 · Updated 2025-12-06

What's in this repo

This page lists the files in martinkorelic/Qwen2-0.5B-MiniRecommendation-mobile-trained, including quant_model.onnx, eval_model.onnx, lora_merger_model.onnx and more.

  • 7 ONNX models (.onnx files)
  • 362 files, including 168 .tensor, 168 .npz, 14 .json, 7 .onnx, 1 .md, 1 .data
  • Last updated 2025-12-06

ONNX files

File Size Download
inference/quant_model.onnx — Download
train/eval_model.onnx — Download
train/lora_merger_model.onnx — Download
train/lora_qmerger_model.onnx — Download
train/mars_qmerger_model.onnx — Download
train/optimizer_model.onnx — Download
train/training_model.onnx — Download

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

📱 Qwen2-0.5B Mobile-Finetuned on MiniRecommendation dataset

The first fully fine-tuned language model trained entirely on a mobile device. This repository contains a Qwen2-0.5B model that was completely fine-tuned on a Google Pixel 6 (8GB RAM). The model was trained by using MobileTransformers on-device LLM PEFT framework for fine-tuning and inference.

📱 Training Setup

  • Device: Google Pixel 6 with 8GB RAM
  • Method: MARS OPT0 (our novel parameter-efficient fine-tuning approach)
    • Rank: r=8
    • Alpha: a=2
  • Framework: MobileTransformers
  • Base Model: Qwen2-0.5B

📊 Dataset

MiniRecommendation - A personalized smartphone automation dataset where users express intents and the model recommends appropriate automatic actions to perform on the device. This task-oriented dataset is specifically designed for on-device intelligence scenarios.

Example interaction:

  • User: "I'm heading home"
  • Model: Recommends turning on navigation, adjusting smart home settings, or sending arrival notifications

🔧 On-device examples

This example shows how a base model can be fine-tuned entirely on-device, meaning no data ever leaves the device. During the process, adapters are trained locally, then merged and integrated into the base model on the mobile phone to produce the final fine-tuned version .

🧩 Base Model ⚙️ On-device Fine-tuned model
MinIRecommendationWrong MiniRecommendationRight

📈 On-device metrics over time

on_device_training_metrics_recommendation-1

📁 Repository Structure

├── outputs/              # Inference outputs from base and fine-tuned models
├── merged-npz/           # Merged adapter weights with base model (npz format)
├── tokenizer/            # Tokenizer artifacts
├── inference/            # Inference artifacts
├── inference/merged/     # Merged inference weights ready to be inserted into the base model
└── train/                # Training artifacts

💡 Why This Matters

Training LLMs on mobile devices opens up:

  • Privacy-first ML: Keep sensitive training data on-device
  • Edge AI democratization: No cloud infrastructure required
  • Personalization: Fine-tune models directly on user devices
  • Accessibility: Enable AI development in resource-constrained environments

Acknowledgements

This work was supported by the Slovenian Research Agency grant no. N2-0393 approXimation for adaptable diStributed artificial intelligence and grant no. J2-3047 Context-Aware On-Device Approximate Computing.

Model info

onnx dataset:martinkorelic/MiniRecommendation base_model:Qwen/Qwen2-0.5B base_model:quantized:Qwen/Qwen2-0.5B region:us
↓ 0 ♥ 0

Created: 2025-10-29

Updated: 2025-12-06

View on Hugging Face

Files (362)

.gitattributes
README.md
inference/generation_config.json
inference/merged/model.layers.0.attn.k_proj.MatMul.tensor
inference/merged/model.layers.0.attn.o_proj.MatMul.tensor
inference/merged/model.layers.0.attn.q_proj.MatMul.tensor
inference/merged/model.layers.0.attn.v_proj.MatMul.tensor
inference/merged/model.layers.0.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.0.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.0.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.1.attn.k_proj.MatMul.tensor
inference/merged/model.layers.1.attn.o_proj.MatMul.tensor
inference/merged/model.layers.1.attn.q_proj.MatMul.tensor
inference/merged/model.layers.1.attn.v_proj.MatMul.tensor
inference/merged/model.layers.1.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.1.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.1.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.10.attn.k_proj.MatMul.tensor
inference/merged/model.layers.10.attn.o_proj.MatMul.tensor
inference/merged/model.layers.10.attn.q_proj.MatMul.tensor
inference/merged/model.layers.10.attn.v_proj.MatMul.tensor
inference/merged/model.layers.10.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.10.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.10.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.11.attn.k_proj.MatMul.tensor
inference/merged/model.layers.11.attn.o_proj.MatMul.tensor
inference/merged/model.layers.11.attn.q_proj.MatMul.tensor
inference/merged/model.layers.11.attn.v_proj.MatMul.tensor
inference/merged/model.layers.11.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.11.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.11.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.12.attn.k_proj.MatMul.tensor
inference/merged/model.layers.12.attn.o_proj.MatMul.tensor
inference/merged/model.layers.12.attn.q_proj.MatMul.tensor
inference/merged/model.layers.12.attn.v_proj.MatMul.tensor
inference/merged/model.layers.12.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.12.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.12.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.13.attn.k_proj.MatMul.tensor
inference/merged/model.layers.13.attn.o_proj.MatMul.tensor
inference/merged/model.layers.13.attn.q_proj.MatMul.tensor
inference/merged/model.layers.13.attn.v_proj.MatMul.tensor
inference/merged/model.layers.13.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.13.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.13.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.14.attn.k_proj.MatMul.tensor
inference/merged/model.layers.14.attn.o_proj.MatMul.tensor
inference/merged/model.layers.14.attn.q_proj.MatMul.tensor
inference/merged/model.layers.14.attn.v_proj.MatMul.tensor
inference/merged/model.layers.14.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.14.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.14.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.15.attn.k_proj.MatMul.tensor
inference/merged/model.layers.15.attn.o_proj.MatMul.tensor
inference/merged/model.layers.15.attn.q_proj.MatMul.tensor
inference/merged/model.layers.15.attn.v_proj.MatMul.tensor
inference/merged/model.layers.15.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.15.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.15.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.16.attn.k_proj.MatMul.tensor
inference/merged/model.layers.16.attn.o_proj.MatMul.tensor
inference/merged/model.layers.16.attn.q_proj.MatMul.tensor
inference/merged/model.layers.16.attn.v_proj.MatMul.tensor
inference/merged/model.layers.16.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.16.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.16.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.17.attn.k_proj.MatMul.tensor
inference/merged/model.layers.17.attn.o_proj.MatMul.tensor
inference/merged/model.layers.17.attn.q_proj.MatMul.tensor
inference/merged/model.layers.17.attn.v_proj.MatMul.tensor
inference/merged/model.layers.17.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.17.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.17.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.18.attn.k_proj.MatMul.tensor
inference/merged/model.layers.18.attn.o_proj.MatMul.tensor
inference/merged/model.layers.18.attn.q_proj.MatMul.tensor
inference/merged/model.layers.18.attn.v_proj.MatMul.tensor
inference/merged/model.layers.18.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.18.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.18.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.19.attn.k_proj.MatMul.tensor
inference/merged/model.layers.19.attn.o_proj.MatMul.tensor
inference/merged/model.layers.19.attn.q_proj.MatMul.tensor
inference/merged/model.layers.19.attn.v_proj.MatMul.tensor
inference/merged/model.layers.19.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.19.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.19.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.2.attn.k_proj.MatMul.tensor
inference/merged/model.layers.2.attn.o_proj.MatMul.tensor
inference/merged/model.layers.2.attn.q_proj.MatMul.tensor
inference/merged/model.layers.2.attn.v_proj.MatMul.tensor
inference/merged/model.layers.2.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.2.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.2.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.20.attn.k_proj.MatMul.tensor
inference/merged/model.layers.20.attn.o_proj.MatMul.tensor
inference/merged/model.layers.20.attn.q_proj.MatMul.tensor
inference/merged/model.layers.20.attn.v_proj.MatMul.tensor
inference/merged/model.layers.20.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.20.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.20.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.21.attn.k_proj.MatMul.tensor
inference/merged/model.layers.21.attn.o_proj.MatMul.tensor
inference/merged/model.layers.21.attn.q_proj.MatMul.tensor
inference/merged/model.layers.21.attn.v_proj.MatMul.tensor
inference/merged/model.layers.21.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.21.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.21.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.22.attn.k_proj.MatMul.tensor
inference/merged/model.layers.22.attn.o_proj.MatMul.tensor
inference/merged/model.layers.22.attn.q_proj.MatMul.tensor
inference/merged/model.layers.22.attn.v_proj.MatMul.tensor
inference/merged/model.layers.22.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.22.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.22.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.23.attn.k_proj.MatMul.tensor
inference/merged/model.layers.23.attn.o_proj.MatMul.tensor
inference/merged/model.layers.23.attn.q_proj.MatMul.tensor
inference/merged/model.layers.23.attn.v_proj.MatMul.tensor
inference/merged/model.layers.23.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.23.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.23.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.3.attn.k_proj.MatMul.tensor
inference/merged/model.layers.3.attn.o_proj.MatMul.tensor
inference/merged/model.layers.3.attn.q_proj.MatMul.tensor
inference/merged/model.layers.3.attn.v_proj.MatMul.tensor
inference/merged/model.layers.3.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.3.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.3.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.4.attn.k_proj.MatMul.tensor
inference/merged/model.layers.4.attn.o_proj.MatMul.tensor
inference/merged/model.layers.4.attn.q_proj.MatMul.tensor
inference/merged/model.layers.4.attn.v_proj.MatMul.tensor
inference/merged/model.layers.4.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.4.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.4.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.5.attn.k_proj.MatMul.tensor
inference/merged/model.layers.5.attn.o_proj.MatMul.tensor
inference/merged/model.layers.5.attn.q_proj.MatMul.tensor
inference/merged/model.layers.5.attn.v_proj.MatMul.tensor
inference/merged/model.layers.5.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.5.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.5.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.6.attn.k_proj.MatMul.tensor
inference/merged/model.layers.6.attn.o_proj.MatMul.tensor
inference/merged/model.layers.6.attn.q_proj.MatMul.tensor
inference/merged/model.layers.6.attn.v_proj.MatMul.tensor
inference/merged/model.layers.6.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.6.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.6.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.7.attn.k_proj.MatMul.tensor
inference/merged/model.layers.7.attn.o_proj.MatMul.tensor
inference/merged/model.layers.7.attn.q_proj.MatMul.tensor
inference/merged/model.layers.7.attn.v_proj.MatMul.tensor
inference/merged/model.layers.7.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.7.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.7.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.8.attn.k_proj.MatMul.tensor
inference/merged/model.layers.8.attn.o_proj.MatMul.tensor
inference/merged/model.layers.8.attn.q_proj.MatMul.tensor
inference/merged/model.layers.8.attn.v_proj.MatMul.tensor
inference/merged/model.layers.8.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.8.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.8.mlp.up_proj.MatMul.tensor
inference/merged/model.layers.9.attn.k_proj.MatMul.tensor
inference/merged/model.layers.9.attn.o_proj.MatMul.tensor
inference/merged/model.layers.9.attn.q_proj.MatMul.tensor
inference/merged/model.layers.9.attn.v_proj.MatMul.tensor
inference/merged/model.layers.9.mlp.down_proj.MatMul.tensor
inference/merged/model.layers.9.mlp.gate_proj.MatMul.tensor
inference/merged/model.layers.9.mlp.up_proj.MatMul.tensor
inference/quant_model.onnx ONNX
inference/quant_model.onnx.data
merged-npz/backbone.model.layers.0.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.0.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.0.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.0.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.0.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.0.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.0.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.1.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.1.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.1.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.1.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.1.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.1.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.1.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.10.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.10.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.10.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.10.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.10.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.10.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.10.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.11.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.11.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.11.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.11.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.11.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.11.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.11.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.12.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.12.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.12.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.12.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.12.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.12.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.12.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.13.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.13.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.13.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.13.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.13.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.13.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.13.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.14.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.14.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.14.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.14.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.14.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.14.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.14.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.15.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.15.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.15.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.15.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.15.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.15.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.15.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.16.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.16.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.16.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.16.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.16.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.16.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.16.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.17.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.17.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.17.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.17.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.17.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.17.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.17.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.18.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.18.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.18.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.18.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.18.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.18.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.18.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.19.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.19.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.19.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.19.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.19.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.19.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.19.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.2.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.2.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.2.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.2.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.2.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.2.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.2.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.20.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.20.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.20.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.20.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.20.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.20.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.20.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.21.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.21.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.21.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.21.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.21.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.21.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.21.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.22.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.22.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.22.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.22.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.22.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.22.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.22.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.23.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.23.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.23.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.23.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.23.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.23.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.23.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.3.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.3.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.3.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.3.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.3.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.3.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.3.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.4.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.4.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.4.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.4.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.4.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.4.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.4.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.5.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.5.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.5.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.5.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.5.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.5.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.5.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.6.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.6.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.6.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.6.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.6.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.6.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.6.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.7.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.7.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.7.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.7.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.7.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.7.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.7.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.8.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.8.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.8.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.8.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.8.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.8.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.8.self_attn.v_proj.base_layer.npz
merged-npz/backbone.model.layers.9.mlp.down_proj.base_layer.npz
merged-npz/backbone.model.layers.9.mlp.gate_proj.base_layer.npz
merged-npz/backbone.model.layers.9.mlp.up_proj.base_layer.npz
merged-npz/backbone.model.layers.9.self_attn.k_proj.base_layer.npz
merged-npz/backbone.model.layers.9.self_attn.o_proj.base_layer.npz
merged-npz/backbone.model.layers.9.self_attn.q_proj.base_layer.npz
merged-npz/backbone.model.layers.9.self_attn.v_proj.base_layer.npz
outputs/base_evaluation_outputs.json
outputs/comparison_evaluation_results.json
outputs/finetuned_evaluation_outputs.json
tokenizer/added_tokens.json
tokenizer/merges.txt
tokenizer/ortmobile_tokenizer_config.json
tokenizer/special_tokens_map.json
tokenizer/tokenizer.json
tokenizer/tokenizer_config.json
tokenizer/vocab.json
train/checkpoint
train/eval_model.onnx ONNX
train/lora_merger_model.onnx ONNX
train/lora_qmerger_model.onnx ONNX
train/mars_qmerger_model.onnx ONNX
train/optimizer_model.onnx ONNX
train/training_config.json
train/training_information.json
train/training_logs.json
train/training_model.onnx ONNX
train/training_state.json

Related models