CoolFace
Modelpublic

aitf-komdigi/KomdigiITS-8B-DFK-TextClassification-Final

sourceHugging Faceapache-2.0updated 3mo agoView on Hugging Face
0likes4downloads
Model Card

Dataset Details

Tasks & Capabilities

  • —Text Classification DFK (Disinformasi, Fitnah, Ujaran Kebencian)

Dataset Splits

  • —Train Samples: 20,790
  • —Validation Samples: 2,599
  • —Testing Samples: 2,599

Target Labels

  1. 1.DISINFORMASI
  2. 2.FITNAH
  3. 3.UJARAN KEBENCIAN
  4. 4.FAKTA
  5. 5.BUKAN DFK

Training Configurations

Supervised Fine-Tuning (SFT) Pipeline Parameters

ParameterValue
Max Sequence Length2048 tokens
Batch Size2
Gradient Accumulation Steps16
Effective Batch Size32
Learning Rate (LR)2e-4
LR SchedulerLinear
OptimizerAdamW 8-bit
Number of Epochs1
Warmup Steps20
Weight Decay0.01
Max Gradient Norm1.0
Evaluation StepsEvery 200 steps

LoRA (PEFT) Hyperparameters

ParameterValue
r (Rank)16
lora_alpha32
lora_dropout0
biasnone
use_rsloraFalse
Gradient Checkpointingunsloth
Target Modulesq_proj, k_proj, v_proj, o_proj, gate_proj, down_proj, up_proj

Uploaded model

  • —Developed by: hnuka
  • —License: apache-2.0
  • —Finetuned from model : aitf-komdigi/KomdigiITS-8B-DFK-CPT

This mistral3 model was trained 2x faster with Unsloth

<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>