use warmup_ratio as a better default than warmup steps since it's data dependent (#2897) [skip ci]
* use warmup_ratio as a better default than warmup steps since it's data dependent * replace remainder of warmup_steps
This commit is contained in:
@@ -59,7 +59,7 @@ gradient_checkpointing: true
|
||||
resume_from_checkpoint:
|
||||
logging_steps: 1
|
||||
|
||||
warmup_steps: 10
|
||||
warmup_ratio: 0.1
|
||||
evals_per_epoch: 4
|
||||
saves_per_epoch: 4
|
||||
weight_decay: 0.0
|
||||
|
||||
@@ -50,7 +50,7 @@ resume_from_checkpoint:
|
||||
logging_steps: 1
|
||||
flash_attention: true
|
||||
|
||||
warmup_steps: 100
|
||||
warmup_ratio: 0.1
|
||||
evals_per_epoch: 4
|
||||
saves_per_epoch: 1
|
||||
weight_decay: 0.1
|
||||
|
||||
@@ -53,7 +53,7 @@ resume_from_checkpoint:
|
||||
logging_steps: 1
|
||||
flash_attention: true
|
||||
|
||||
warmup_steps: 100
|
||||
warmup_ratio: 0.1
|
||||
evals_per_epoch: 4
|
||||
saves_per_epoch: 1
|
||||
weight_decay: 0.1
|
||||
|
||||
@@ -50,7 +50,7 @@ resume_from_checkpoint:
|
||||
logging_steps: 1
|
||||
flash_attention: true
|
||||
|
||||
warmup_steps: 100
|
||||
warmup_ratio: 0.1
|
||||
evals_per_epoch: 4
|
||||
saves_per_epoch: 1
|
||||
weight_decay: 0.1
|
||||
|
||||
@@ -51,7 +51,7 @@ resume_from_checkpoint:
|
||||
logging_steps: 1
|
||||
flash_attention: true
|
||||
|
||||
warmup_steps: 100
|
||||
warmup_ratio: 0.1
|
||||
evals_per_epoch: 4
|
||||
saves_per_epoch: 1
|
||||
weight_decay: 0.1
|
||||
|
||||
Reference in New Issue
Block a user