Greninja9257/LabLLM
build the architecture, train the weights, and watch a small LLM emerge from scratch, locally on Apple Silicon with custom data, tokenizers, checkpoints, and MLX acceleration.
在 GitHub 打开每天早上一封邮件，把昨天的 AI 梳理好订阅邮件
每天早上一封邮件，把昨天的 AI 梳理好订阅邮件
build the architecture, train the weights, and watch a small LLM emerge from scratch, locally on Apple Silicon with custom data, tokenizers, checkpoints, and MLX acceleration.
在 GitHub 打开