We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent f834263 commit c5db813Copy full SHA for c5db813
recipes/configs/qwen2/7B_full_single_device.yaml
@@ -60,7 +60,7 @@ optimizer_in_bwd: True
60
loss:
61
_component_: torchtune.modules.loss.CEWithChunkedOutputLoss
62
max_steps_per_epoch: null
63
-gradient_accumulation_steps: 16
+gradient_accumulation_steps: 1
64
compile: False
65
66
# Training environment
0 commit comments