root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 02:44:36 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

readme : add some recent perplexity and bpw measurements to READMES, link for k-quants (#3340)

Browse Source 
* Update README.md

* Update README.md

* Update README.md with k-quants bpw measurements
This commit is contained in:
BarfingLemurs 2023-09-27 11:30:36 -04:00 committed by GitHub
parent 99115f3fa6
commit ffe88a36a9
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 64 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
README.md
View File

@ -597,6 +597,11 @@ Several quantization methods are supported. They differ in the resulting model d
| 13B | ms/tok @ 8th | - | 73 | 82 | 98 | 105 | 128 | | 13B | ms/tok @ 8th | - | 73 | 82 | 98 | 105 | 128 |
| 13B | bits/weight | 16.0 | 4.5 | 5.0 | 5.5 | 6.0 | 8.5 | | 13B | bits/weight | 16.0 | 4.5 | 5.0 | 5.5 | 6.0 | 8.5 |
- [k-quants](https://github.com/ggerganov/llama.cpp/pull/1684)
- recent k-quants improvements
- [#2707](https://github.com/ggerganov/llama.cpp/pull/2707)
- [#2807](https://github.com/ggerganov/llama.cpp/pull/2807)
### Perplexity (measuring model quality) ### Perplexity (measuring model quality)
You can use the `perplexity` example to measure perplexity over a given prompt (lower perplexity is better). You can use the `perplexity` example to measure perplexity over a given prompt (lower perplexity is better).

18
examples/perplexity/README.md
View File

@ -1,3 +1,21 @@
# perplexity # perplexity
TODO TODO
## Llama 2 70B Scorechart
Quantization | Model size (GiB) | Perplexity | Delta to fp16
-- | -- | -- | --
Q4_0 | 36.20 | 3.5550 | 3.61%
Q4_1 | 40.20 | 3.5125 | 2.37%
Q5_0 | 44.20 | 3.4744 | 1.26%
Q2_K | 27.27 | 3.7339 | 8.82%
Q3_K_S | 27.86 | 3.7019 | 7.89%
Q3_K_M | 30.83 | 3.5932 | 4.72%
Q3_K_L | 33.67 | 3.5617 | 3.80%
Q4_K_S | 36.39 | 3.4852 | 1.57%
Q4_K_M | 38.54 | 3.4725 | 1.20%
Q5_K_S | 44.20 | 3.4483 | 0.50%
Q5_K_M | 45.41 | 3.4451 | 0.40%
Q6_K | 52.70 | 3.4367 | 0.16%
fp16 | 128.5 | 3.4313 | -

41
examples/quantize/README.md
View File

@ -1,3 +1,44 @@
# quantize # quantize
TODO 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