Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion torchao/quantization/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@ Typically quantization algorithms will have different schemes for how the activa
All the following benchmarks are for `meta-llama/Llama-3-8.1B` using `lm-eval` measured on an H100 GPU.

| weight | activation | wikitext-perplexity | winogrande | checkpoint size (GB) |
| --------- | ------------------- | ---------- | -------------------- |
| --------- | ------------------- | ---------- | -------------------- | -------- |
| bfloat16 | bfloat16 | 7.3315 | 0.7380 | 16.1 |
| float8_rowwise | float8_rowwise | 7.4197 | 0.7388 | 9.1 |
| int8_rowwise | bfloat16 | 7.3451 | 0.7340 | 9.1 |
Expand Down
Loading