Shubhamchoudhari00072's workspace
Runs
4
State
Notes
User
Tags
Created
Runtime
Sweep
batch_size
compile_model
d_model
dataset_name
eval_iters
learning_rate
max_iters
micro_batch
min_learning_rate
multiple_of
num_heads
num_layers
resume_traning
save_ckpt_iters
seq_len
tokenizer_name
warmup_iters
iter
lr
test_loss
time
tokens
trainer/global_step
traning_loss
Killed
shubhamchoudhari00072
23m 19s
-
32
true
256
roneneldan/TinyStories
100
0.0005
100000
4
0
4
8
8
false
1000
256
microsoft/phi-2
1000
3900
0.00049894
1.33642
0.30591
127296000
3900
1.47502
Killed
shubhamchoudhari00072
1m 24s
-
32
true
128
roneneldan/TinyStories
100
0.0005
100000
4
0
4
4
4
false
1000
256
microsoft/phi-2
1000
100
0.00005
9.97852
37.32363
3264000
100
10.02246
Killed
shubhamchoudhari00072
1m 8s
-
32
true
64
roneneldan/TinyStories
100
0.0005
10000
4
0
4
4
4
false
1000
256
microsoft/phi-2
100
100
0.0005
7.70796
35.14156
3264000
100
7.76911
Failed
shubhamchoudhari00072
5s
-
32
true
64
roneneldan/TinyStories
100
0.0005
10000
4
0
4
4
4
false
1000
256
microsoft/phi-2
1000
-
-
-
-
-
-
-
1-4
of 4