Mirrors/llama.cpp

mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 13:58:46 +01:00

History

Justine Tunney 3855416027 ggml : introduce bfloat16 support (#6412 ) * Introduce bfloat16 support Many models on Hugging Face (e.g. Mistral, TinyLLaMA) use bfloat16 as their canonical floating point format. ┌sign │ │ ┌exponent │ │ │ │ ┌mantissa │ │ │ │┌──┴───┐┌─┴───┐ 0b0000000000000000 brain16 This encoding has the same number of exponent bits as float32. That makes conversion relatively straightforward, even in the absence of hardware support. For example, converting brain16 to binary32 means simply shifting 16 bits to the left. ┌sign │ │ ┌exponent │ │ │ │ ┌mantissa │ │ │ │┌──┴───┐┌─┴───────────────────┐ 0b00000000000000000000000000000000 IEEE binary32 The issue is that converting bf16 to fp16 can result in information loss. Only 13% of bf16 numbers can be precisely represented in fp16 which in practice ends up being 99.71% of Mistral 7b v0.2's weights however there is currently no way other than fp32 to get the others ┌sign │ │ ┌exponent │ │ │ │ ┌mantissa │ │ │ │┌─┴─┐┌─┴──────┐ 0b0000000000000000 IEEE binary16 This change fixes that, by adding a bf16 data type to GGML. Support for CPU inference has been implemented along with optimizations for the AVX2, AVX512, and AVX512BF16 ISAs. Perplexity on Mistral 7b 0.2 improves somewhere around -0.0024 to -0.0046 compared to using fp16 * Remove GGML code that's not needed * Minimize the GGML API surface area for BF16 * Remove bf16 luts * Make the GGML header look nicer * Fix documentation * Apply ggerganov's fixes for test-backend-ops * Add BF16 code for new ggml_validate_row_data() function		2024-05-08 09:30:09 +03:00
..
CMakeLists.txt	quantize: add imatrix and dataset metadata in GGUF (#6658 )	2024-04-26 20:06:33 +02:00
quantize.cpp	ggml : introduce bfloat16 support (#6412 )	2024-05-08 09:30:09 +03:00
README.md	chore: Fix markdown warnings (#6625 )	2024-04-12 10:52:36 +02:00
tests.sh	tests : minor bash stuff (#6902 )	2024-04-25 14:27:20 +03:00

README.md

quantize

TODO

Llama 2 7B

Quantization	Bits per Weight (BPW)
Q2_K	3.35
Q3_K_S	3.50
Q3_K_M	3.91
Q3_K_L	4.27
Q4_K_S	4.58
Q4_K_M	4.84
Q5_K_S	5.52
Q5_K_M	5.68
Q6_K	6.56

Llama 2 13B

Quantization	Bits per Weight (BPW)
Q2_K	3.34
Q3_K_S	3.48
Q3_K_M	3.89
Q3_K_L	4.26
Q4_K_S	4.56
Q4_K_M	4.83
Q5_K_S	5.51
Q5_K_M	5.67
Q6_K	6.56

Llama 2 70B

Quantization	Bits per Weight (BPW)
Q2_K	3.40
Q3_K_S	3.47
Q3_K_M	3.85
Q3_K_L	4.19
Q4_K_S	4.53
Q4_K_M	4.80
Q5_K_S	5.50
Q5_K_M	5.65
Q6_K	6.56