Howuhh's group workspace
Group: sweep_v0-v0
Name
114 visualized
alpha: null
alpha: null
114
State
Notes
User
Tags
Created
Runtime
Sweep
batch_size
character
checkpoints_path
data_mode
eval_episodes
eval_every
eval_processes
eval_seed
group
learning_rate
mlc_job_name
name
project
rnn_dropout
rnn_hidden_dim
rnn_layers
seq_len
train_seed
update_steps
use_prev_action
version
weight_decay
alpha
clip_range
gamma
tau
num_heads
expectile_tau
temperature
depth_max
depth_mean
depth_median
depth_min
depth_std
loss
reward_max
reward_mean
reward_median
reward_min
reward_std
times/backward_pass
times/batch_loading_cpu
times/batch_loading_gpu
times/evaluation_cpu
Failed
-
howuhh
2d 1h 35m 10s
64
["sam-hum-law","tou-hum-neu","val-dwa-law","val-hum-law","val-hum-neu","wiz-elf-cha","wiz-gno-neu","wiz-hum-cha","wiz-hum-neu","wiz-orc-cha"]
checkpoints
memmap
50
10000
14
50
sweep_v0
0.0003
["selectel-a100-1x-katakomba-4owoes","selectel-a100-1x-katakomba-644y8i","selectel-a100-1x-katakomba-6mota1","selectel-a100-1x-katakomba-atuky7","selectel-a100-1x-katakomba-ehnqlx","selectel-a100-1x-katakomba-r4h76n","selectel-a100-1x-katakomba-sb17k7","selectel-a100-1x-katakomba-t3l4yv","selectel-a100-1x-katakomba-vz58go","selectel-a100-1x-katakomba-y7bb8t"]
["bc-wiz-elf-cha-10c3f3bf","bc-wiz-elf-cha-5c390de9","bc-wiz-elf-cha-5dd9c308","bc-wiz-gno-neu-64b8d0df","bc-wiz-gno-neu-9ef2d730","bc-wiz-gno-neu-fa02ec07","bc-wiz-hum-cha-481ab0b8","bc-wiz-orc-cha-0113573b","bc-wiz-orc-cha-4182a29b","bc-wiz-orc-cha-8a033b90"]
NetHack
0
2048
2
16
1
500000
true
v0
0
-
-
-
-
-
-
-
0.75439
0.025614
0
0
0.12791
0.24066
2441.08772
512.30298
353.01754
2.37719
532.01501
0.011498
0.020324
0.020299
100.13609
1-1
of 1