extensions/csrc/kernel/cuda - hpcaitech/ColossalAI

Making large AI models cheaper, faster and more accessible

0 0 38 Python

	attention
	utils
	activation_kernel.cu	2.6 KB
	context_kv_cache_memcpy_kernel.cu	10.5 KB
	convert_fp8_kernel.cu	3.8 KB
	decode_kv_cache_memcpy_kernel.cu	9.8 KB
	flash_decoding_attention_kernel.cu	40.5 KB
	fused_rotary_emb_and_cache_kernel.cu	21.3 KB
	get_cos_and_sin_kernel.cu	9.6 KB
	layer_norm_kernel.cu	25.2 KB
	moe_kernel.cu	25.3 KB
	multi_tensor_adam_kernel.cu	4.9 KB
	multi_tensor_apply.cuh	5.1 KB
	multi_tensor_l2norm_kernel.cu	15.0 KB
	multi_tensor_lamb_kernel.cu	12.8 KB
	multi_tensor_scale_kernel.cu	4.3 KB
	multi_tensor_sgd_kernel.cu	6.3 KB
	rms_layernorm_kernel.cu	13.8 KB
	scaled_masked_softmax_kernel.cu	21.6 KB
	scaled_upper_triang_masked_softmax_kernel.cu	22.9 KB