423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2026-01-13 09:30:34 +08:00
Code Issues Packages Projects Releases Wiki Activity

[fix] correct ktransformers example config paths and templates (#9732)

Browse Source
This commit is contained in:
Peilin Li
2026-01-08 10:52:50 +08:00
committed by GitHub
parent 5cfd804b59
commit 03a70ba8dd
7 changed files with 12 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/ktransformers/train_lora/deepseek2_lora_sft_kt.yaml
View File

@@ -10,7 +10,7 @@ lora_rank: 8
lora_target: all
### dataset
dataset: identity
dataset: identity, alpaca_en_demo
template: deepseek
cutoff_len: 2048
max_samples: 100000
@@ -40,7 +40,7 @@ resume_from_checkpoint: null
### ktransformers
use_kt: true # use KTransformers as LoRA sft backend
kt_optimize_rule: examples/kt_optimize_rules/DeepSeek-V2-Lite-Chat-sft-amx.yaml
kt_optimize_rule: examples/ktransformers/kt_optimize_rules/DeepSeek-V2-Lite-Chat-sft-amx.yaml
cpu_infer: 32
chunk_size: 8192

6
examples/ktransformers/train_lora/deepseek3_lora_sft_kt.yaml
View File

@@ -10,8 +10,8 @@ lora_rank: 8
lora_target: all
### dataset
dataset: identity
template: deepseek
dataset: identity, alpaca_en_demo
template: deepseek3
cutoff_len: 2048
max_samples: 100000
overwrite_cache: true
@@ -40,7 +40,7 @@ resume_from_checkpoint: null
### ktransformers
use_kt: true # use KTransformers as LoRA sft backend
kt_optimize_rule: examples/kt_optimize_rules/DeepSeek-V3-Chat-sft-amx-multi-gpu.yaml
kt_optimize_rule: examples/ktransformers/kt_optimize_rules/DeepSeek-V3-Chat-sft-amx-multi-gpu.yaml
cpu_infer: 32
chunk_size: 8192

2
examples/ktransformers/train_lora/qwen3moe_lora_sft_kt.yaml
View File

@@ -40,7 +40,7 @@ resume_from_checkpoint: null
### ktransformers
use_kt: true # use KTransformers as LoRA sft backend
kt_optimize_rule: examples/kt_optimize_rules/Qwen3Moe-sft-amx.yaml
kt_optimize_rule: examples/ktransformers/kt_optimize_rules/Qwen3Moe-sft-amx.yaml
cpu_infer: 32
chunk_size: 8192