pre-commits

This commit is contained in:
Sotirios Anagnostidis
2023-01-03 22:45:34 +01:00
parent 752eeb24f7
commit c20dfaad5b
4 changed files with 17 additions and 11 deletions
@@ -32,6 +32,17 @@ galactica-125:
per_device_train_batch_size: 4
per_device_eval_batch_size: 4
gpt-jt:
learning_rate: 2e-6
model_name: togethercomputer/GPT-JT-6B-v1
weight_decay: 0.01
max_length: 1024
warmup_steps: 600
gradient_checkpointing: false
gradient_accumulation_steps: 2
per_device_train_batch_size: 4
per_device_eval_batch_size: 4
debug:
eval_steps: 20
eval_size: 100