GitHub - adengitstuff/llm_created: Building a large model from scratch, following Raschka's book! Extending this for fun to multimodal on tiny, 32x32 character images.

Name		Name	Last commit message	Last commit date
__pycache__		__pycache__
character_images_from_scratch		character_images_from_scratch
model_tests		model_tests
.gitignore		.gitignore
512_test.py		512_test.py
architecture.py		architecture.py
causal_attention_mask.py		causal_attention_mask.py
causalattention.py		causalattention.py
causalmultiatt.py		causalmultiatt.py
characterDataset.py		characterDataset.py
compact_self_attention_class.py		compact_self_attention_class.py
encodetest.py		encodetest.py
first_generated_character_pixel.png		first_generated_character_pixel.png
gelu.py		gelu.py
generate_image_from_scratch.py		generate_image_from_scratch.py
gptpyweights.py		gptpyweights.py
image_data.py		image_data.py
image_model.py		image_model.py
layernorm.py		layernorm.py
nextstep.py		nextstep.py
parallel_multi_attention.py		parallel_multi_attention.py
placeholder_gpt.py		placeholder_gpt.py
real_self_attention.py		real_self_attention.py
self_attention_linear_layers.py		self_attention_linear_layers.py
simple_attention.py		simple_attention.py
sliding_tensor_dataloader.py		sliding_tensor_dataloader.py
the-verdict.txt		the-verdict.txt
transformerblock.py		transformerblock.py

About

Building a large model from scratch, following Raschka's book! Extending this for fun to multimodal on tiny, 32x32 character images.

0 stars

0 watching