Universal Model Trainer

0

Total Jobs

0

Active Jobs

0

Completed

0%

Success Rate

Current Training

No active training. Start a new training job!

System Status

CPU 0%

Memory 0%

Disk 0%

GPU: Checking...

Cache: 0 MB

Configure New Training Job

1

Task

2

Model

3

Dataset

4

Columns

5

Train

Select Task Type

Causal LM

Text Generation

Seq2Seq

Summarization/Translation

Token Classification

NER

Text Classification

Sentiment/Categories

Question Answering

Extractive QA

Reasoning

Chain-of-Thought

Select Model

Select Dataset

Training Configuration

Job Name

Epochs

Batch Size

Learning Rate

Max Sequence Length

128 512 4096

Warmup Steps

PEFT/LoRA Settings

Enable PEFT/LoRA (Recommended for CPU training)

Method

LoRA Rank (r)

LoRA Alpha

LoRA Dropout

Output Settings

Output Model Name

Push to HuggingFace Hub after training

Training Jobs

Browse Models

Browse Datasets

Settings

Configure your training environment

HuggingFace Token Required for pushing models to Hub and accessing gated models

WANDB API Key (Optional)

Train Any Model

0

0

0

0%