diff options
Diffstat (limited to 'candle-transformers/src/models/bigcode.rs')
| -rw-r--r-- | candle-transformers/src/models/bigcode.rs | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/candle-transformers/src/models/bigcode.rs b/candle-transformers/src/models/bigcode.rs index f6b4a4ef..8ed1462b 100644 --- a/candle-transformers/src/models/bigcode.rs +++ b/candle-transformers/src/models/bigcode.rs @@ -1,3 +1,10 @@ +//! BigCode implementation in Rust based on the GPT-BigCode model. +//! +//! See "StarCoder: A State-of-the-Art LLM for Code", Mukherjee et al. 2023 +//! - [Arxiv](https://arxiv.org/abs/2305.06161) +//! - [Github](https://github.com/bigcode-project/starcoder) +//! + use candle::{DType, Device, IndexOp, Result, Tensor, D}; use candle_nn::{embedding, linear_b as linear, Embedding, LayerNorm, Linear, Module, VarBuilder}; |