v6 re-export: legacy torch.onnx.export (opset 14) — fp32 + int8

update tokenizer_config.json

v6: deep LoRA r=64 on train_v2.jsonl (Haiku-annotated topics) — int8 ONNX

Upload tokenizer_config.json with huggingface_hub

Upload onnx/model_quantized.onnx with huggingface_hub

Upload onnx/model_quantized.onnx with huggingface_hub

Delete onnx/decoder_model_merged_quantized.onnx with huggingface_hub

Upload onnx/model_q8.onnx with huggingface_hub

Delete onnx/model_quantized.onnx with huggingface_hub

Upload onnx/decoder_model_merged_quantized.onnx with huggingface_hub

Upload vocab.json with huggingface_hub

Upload tokenizer_config.json with huggingface_hub

Upload tokenizer.json with huggingface_hub

Upload special_tokens_map.json with huggingface_hub

Upload onnx/model_quantized.onnx with huggingface_hub

Upload merges.txt with huggingface_hub

Upload generation_config.json with huggingface_hub

Upload config.json with huggingface_hub

initial commit