Chenmientan's workspace
Runs
1
Name
1 visualized
State
Notes
User
Tags
Created
Runtime
Sweep
critic.fsdp_size
critic.gradient_checkpointing
critic.lr
critic.max_grad_norm
critic.max_length_per_device
critic.model_name
critic.offload_optimizer
critic.save_dir
critic.save_optimizer
critic.sp_size
critic.warmup_ratio
critic.weight_decay
data.batch_size
data.max_length
data.path
trainer.disable_wandb
trainer.experiment_name
trainer.n_epochs
trainer.project
accuray
grad_norm
loss
timing/update_critic
Failed
-
chenmientan
1h 5m 24s
-
0
true
0.00001
1
4096
/root/.cache/huggingface/hub/models--meta-llama--Llama-3.1-8B-Instruct/snapshots/0e9e39f249a16976918f6564b8830bc894c89659
false
ckpts/llama-3.1-8b-inst
true
1
0.1
0.01
128
2048
Chenmien/SkyworkRM
false
llama-3.1-8b-inst
1
SkyworkRM
0.90909
6.67904
0.11631
3.8041
1-1
of 1