jedisct1 commited on
Commit
2e95e78
·
verified ·
1 Parent(s): 41e8397

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -6
README.md CHANGED
@@ -17,7 +17,7 @@ datasets:
17
 
18
  # Qwen3.6-35B-A3B Rust Code Fine-Tune (MLX)
19
 
20
- A fine-tuned version of [Qwen3.6-35B-A3B](https://huggingface.co/Brooooooklyn/Qwen3.6-35B-A3B-UD-Q8_K_XL-mlx) optimized for Rust code generation and comprehension.
21
 
22
  The base model is a Mixture-of-Experts architecture with 35B total parameters and 3B active per token (256 experts, 8 active). Quantized to 8-bit.
23
 
@@ -98,8 +98,4 @@ The fine-tuned model differs from the base model in several ways:
98
 
99
  - Trained with max sequence length of 512 tokens. Patches longer than ~400 tokens of code may be truncated or incomplete.
100
  - The model was trained on commit-level diffs, so it works best for focused, single-purpose changes rather than large refactors spanning many files.
101
- - As with any code generation model, outputs should be reviewed before use. The model may produce syntactically valid but semantically incorrect code.
102
-
103
- ## Training reproduction
104
-
105
- Full training scripts and instructions are available in the [training repository](https://github.com/jedisct1/qwen-rust-finetune). The process requires an Apple Silicon Mac with at least 64 GB unified memory and takes about 90 minutes for 1000 iterations.
 
17
 
18
  # Qwen3.6-35B-A3B Rust Code Fine-Tune (MLX)
19
 
20
+ A fine-tuned version of Qwen3.6-35B-A3B optimized for Rust code generation and comprehension.
21
 
22
  The base model is a Mixture-of-Experts architecture with 35B total parameters and 3B active per token (256 experts, 8 active). Quantized to 8-bit.
23
 
 
98
 
99
  - Trained with max sequence length of 512 tokens. Patches longer than ~400 tokens of code may be truncated or incomplete.
100
  - The model was trained on commit-level diffs, so it works best for focused, single-purpose changes rather than large refactors spanning many files.
101
+ - As with any code generation model, outputs should be reviewed before use. The model may produce syntactically valid but semantically incorrect code.