AshkanTaghipour/GeoLLM-Qwen3.5-4B

Fine-tuned Qwen3.5-4B for mineral exploration geology, targeting the Western Australian geological domain. This model is part of the GeoLLM-Qwen3.5-FineTune benchmark. All five model sizes (0.8B—27B) were trained with identical hyperparameters on the same dataset for fair comparison. — Method: bf16 LoRA (r=16, alpha=16) via Unsloth + SFTTrainer — Epochs: 5 — Optimizer: adamw8bit, cosine LR schedule (2e-4, 10% warmup) — Dataset: mineral-exploration-geology-qa (479 train, 26 test examples) — Hardware: NVIDIA A100-80GB — VRAM (training): 10 GB — VRAM (inference):** ~9 GB

Модальности:
Генерация текста

Области применения:
Диалог / чат

Задача: Генерация текста
Автор: AshkanTaghipour
Теги: qwen3_5, geology, mineral-exploration, qwen3.5, lora, fine-tuned, geoscience, conversational
Лайков: 3  |  Загрузок: 22

Открыть на HuggingFace →

Описание основано на материалах HuggingFace. Перевод выполнен автоматически.