LLM.int8() performs 8-bit inference for transformers up to 175B parameters with no accuracy loss by combining vector-wise quantization for most features with 16-bit mixed-precision handling of systematic outlier dimensions.
arXiv preprint arXiv:1907.04840 , year=
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
An adaptive regularization update for Bregman optimizers achieves target sparsity levels from 75% to 99% with faster early convergence and performance matching or exceeding oracle-tuned baselines.
A stochastic column-block nonlinear Bregman method is introduced for sparse solutions of nonlinear systems, with a proven convergence rate bound under stated assumptions.
citing papers explorer
-
LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale
LLM.int8() performs 8-bit inference for transformers up to 175B parameters with no accuracy loss by combining vector-wise quantization for most features with 16-bit mixed-precision handling of systematic outlier dimensions.
-
Adaptive Regularization for Sparsity Control in Bregman-Based Optimizers
An adaptive regularization update for Bregman optimizers achieves target sparsity levels from 75% to 99% with faster early convergence and performance matching or exceeding oracle-tuned baselines.
-
On a stochastic column-block bregman method for nonlinear systems
A stochastic column-block nonlinear Bregman method is introduced for sparse solutions of nonlinear systems, with a proven convergence rate bound under stated assumptions.