Shubhamchoudhari00072's workspace
Runs
12
Name
12 visualized
State
Notes
User
Tags
Created
Runtime
Sweep
BG
always_save_checkpoint
batch_size
beta1
beta2
compile
decay_lr
device
dim
dropout
dtype
eval_interval
eval_iters
eval_only
grad_clip
gradient_accumulation_steps
init_from
learning_rate
log_interval
max_iters
max_seq_len
multiple_of
n_heads
n_kv_heads
n_layers
out_dir
vocab_size
vocab_source
wandb_log
wandb_project
wandb_run_name
warmup_iters
weight_decay
iter
loss/train
loss/val
lr
mfu
tokens
Finished
-
shubhamchoudhari00072
4h 13m
-
-
true
16
0.9
0.95
true
true
cuda
512
0
float32
200
100
false
1
4
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_25_10_44_19
1000
0.1
16000
1.12396
1.10768
0.00047221
3.34068
524288000
Crashed
-
shubhamchoudhari00072
2h 4m 38s
-
-
true
16
0.9
0.95
true
true
cuda
512
0
float32
200
100
false
1
4
scratch
0.0005
1
16010
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_25_07_43_56
1000
0.1
7800
1.16543
1.14839
0.00028676
3.36566
255590400
Killed
-
shubhamchoudhari00072
ghostmax
4h 10m 15s
-
-
true
16
0.9
0.95
true
true
cuda
512
0
float32
200
100
false
1
4
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_25_03_22_53
1000
0.1
16000
1.12405
1.10735
0.00047221
3.36414
524288000
Crashed
-
shubhamchoudhari00072
3h 22m 34s
-
-
true
16
0.9
0.95
true
true
cuda
512
0
float32
200
100
false
1
4
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_16_38_51
1000
0.1
12800
1.14393
1.12491
0.00048268
3.33207
419430400
Killed
-
shubhamchoudhari00072
6m 55s
-
-
true
16
0.9
0.95
true
true
cuda
512
0
float32
200
100
false
1
4
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_16_17_47
1000
0.1
200
3.91725
3.90338
0.0001
2.04066
6553600
Killed
-
shubhamchoudhari00072
29s
-
-
true
16
0.9
0.95
true
true
cuda
512
0
float32
2000
100
false
1
4
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_16_16_06
1000
0.1
-
-
-
-
-
-
Killed
-
shubhamchoudhari00072
7m 44s
-
-
true
16
0.9
0.95
true
true
cuda
512
0
float32
2000
100
false
1
4
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_16_07_35
1000
0.1
0
10.5013
10.50341
0
-100
0
Killed
-
shubhamchoudhari00072
11m 35s
-
6
true
21
0.9
0.95
true
true
cuda
512
0
float32
2000
100
false
1
24
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_15_52_09
1000
0.1
0
10.50232
10.50295
0
-100
0
Failed
-
shubhamchoudhari00072
1m 28s
-
4
true
32
0.9
0.95
true
true
cuda
512
0
float32
2000
100
false
1
16
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_15_50_20
1000
0.1
0
10.50196
10.50265
0
-100
0
Crashed
-
shubhamchoudhari00072
2m 37s
-
8
true
16
0.9
0.95
false
true
cuda
512
0
float32
2000
100
false
1
32
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_15_20_21
1000
0.1
0
10.5013
10.50341
0
-100
0
Failed
-
shubhamchoudhari00072
1m 42s
-
4
true
32
0.9
0.95
false
true
cuda
512
0
float32
2000
100
false
1
16
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_15_18_09
1000
0.1
0
10.50196
10.50265
0
-100
0
Failed
-
shubhamchoudhari00072
16s
-
4
true
32
0.9
0.95
true
true
cuda
512
0
float32
2000
100
false
1
16
scratch
0.0005
1
100000
512
32
8
8
8
out
32000
llama2
true
ghostmax
run2023_08_24_15_16_58
1000
0.1
-
-
-
-
-
-
1-12
of 12