Architecture: GPT-like Parameters: ~100M Precision: INT8 Format: safetensors Tokenizer: BPE Status: Initialized (not trained)
Files info