Aariyak's workspace
Runs
3
State
Notes
User
Tags
Created
Runtime
Sweep
batch_max_len
beta1
beta2
data_prefix
deepscale
deepspeed
deepspeed_config
deepspeed_mpi
device
epochs
eps
local_rank
lr
lr_min_ratio
lr_warmup_ratio
model_path
model_type
save_every
save_path
weight_decay
Crashed
-
aariyak
1m
-
32768
0.9
0.95
/home/ubuntu/openchat_moe/MIXTRAL_TOKENIZED/data
false
true
ochat/training_deepspeed_zero3/deepspeed_zero3_config.json
false
cuda:0
5
0.00001
0
0.000023635
0.1
0.05
mistralai/Mixtral-8x7B-v0.1
openchat_3.6
1
/home/ubuntu/trained_models/openchat_3.6
0.1
Crashed
-
aariyak
2m 18s
-
32768
0.9
0.95
/home/ubuntu/openchat_moe/MIXTRAL_TOKENIZED/data
false
true
ochat/training_deepspeed_zero3/deepspeed_zero3_config.json
false
cuda:0
5
0.00001
0
0.000023635
0.1
0.05
mistralai/Mixtral-8x7B-v0.1
openchat_3.6
1
/home/ubuntu/trained_models/openchat_3.6
0.1
Failed
-
aariyak
1m 32s
-
32768
0.9
0.95
/home/ubuntu/openchat_moe/MIXTRAL_TOKENIZED/data
false
true
ochat/training_deepspeed_zero3/deepspeed_zero3_config.json
false
cuda:0
5
0.00001
0
0.000023635
0.1
0.05
mistralai/Mixtral-8x7B-v0.1
openchat_3.6
1
run1
0.1
1-3
of 3