ros/LLaMA-Factory
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[model] add models (#7054)

Browse Source 
* add qwen25vl awq models

* add moonlight

Former-commit-id: ae3be2970fea8a35907202a313ab767381c44916
This commit is contained in:
hoshi-hiyouga
2025-02-24 22:05:13 +08:00
committed by GitHub
parent ee46011b34
commit c1d5073bd3
7 changed files with 44 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
examples/train_full/llama3_full_sft.yaml
View File

@@ -34,6 +34,7 @@ bf16: true
ddp_timeout: 180000000
### eval
# eval_dataset: alpaca_en_demo
# val_size: 0.1
# per_device_eval_batch_size: 1
# eval_strategy: steps

1
examples/train_lora/llama3_lora_dpo.yaml
View File

@@ -37,6 +37,7 @@ bf16: true
ddp_timeout: 180000000
### eval
# eval_dataset: dpo_en_demo
# val_size: 0.1
# per_device_eval_batch_size: 1
# eval_strategy: steps

1
examples/train_lora/llama3_lora_pretrain.yaml
View File

@@ -34,6 +34,7 @@ bf16: true
ddp_timeout: 180000000
### eval
# eval_dataset: c4_demo
# val_size: 0.1
# per_device_eval_batch_size: 1
# eval_strategy: steps

1
examples/train_lora/llama3_lora_reward.yaml
View File

@@ -35,6 +35,7 @@ bf16: true
ddp_timeout: 180000000
### eval
# eval_dataset: dpo_en_demo
# val_size: 0.1
# per_device_eval_batch_size: 1
# eval_strategy: steps

1
examples/train_lora/llama3_lora_sft.yaml
View File

@@ -35,6 +35,7 @@ bf16: true
ddp_timeout: 180000000
### eval
# eval_dataset: alpaca_en_demo
# val_size: 0.1
# per_device_eval_batch_size: 1
# eval_strategy: steps