Araffin's workspace
Runs
1,067
Name
1,067 visualized
State
Notes
User
Tags
Created
Runtime
Sweep
base_model
batch_size
cuda
debug_normalize
deepspeed
distributed
eps
exp_name
global_learner_decices
gradient_accumulation_steps
hf_entity
label_dataset
labels.num_labels
labels.num_train
learner_device_ids
learner_devices
local_batch_size
local_micro_batch_size
local_normalize_samples
local_rank
local_rollout_batch_size
lr
normalize_after
normalize_before
normalize_samples
ppo.batch_size
ppo.cliprange
ppo.cliprange_value
ppo.eps
ppo.gamma
ppo.gradient_accumulation_steps
ppo.lam
ppo.local_batch_size
ppo.local_micro_batch_size
ppo.local_mini_batch_size
ppo.lr
ppo.minibatch_size
ppo.nminibatches
ppo.noptepochs
ppo.num_updates
ppo.total_episodes
ppo.vf_coef
ppo.whiten_rewards
ppo.world_size
Finished
costa-huang
1.0.0b1
gpt2
pr-24
pt_adam
1h 18m 21s
-
gpt2
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
1
0.95
64
64
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2
pr-24
pt_adam
5m 29s
-
gpt2
32
true
0
false
-
0.00001
train_reward_accelerate_pt_adam_gpt2
-
1
-
sentiment/offline_5k.json
4
4992
-
-
4
4
256
-
512
0.00005
true
true
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
7h 40m 2s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
7h 33m 36s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
7h 36m 31s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
tf_adam
7h 53m 28s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_tf_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
7h 38m 33s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
tf_adam
7h 52m 4s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_tf_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
tf_adam
7h 52m 58s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_tf_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
tf_adam
7h 52m 55s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_tf_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
tf_adam
7h 51m 26s
-
gpt2-xl
-
true
-
false
-
-
train_policy_accelerate_tf_adam_gpt2_xl_grad_accu
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
4
0.95
64
16
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2
pr-24
pt_adam
1h 17m 35s
-
gpt2
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
1
0.95
64
64
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2
pr-24
pt_adam
1h 16m 17s
-
gpt2
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
1
0.95
64
64
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2
pr-24
pt_adam
1h 16m 54s
-
gpt2
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
1
0.95
64
64
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2
pr-24
pt_adam
1h 16m 24s
-
gpt2
-
true
-
false
-
-
train_policy_accelerate_pt_adam_gpt2
-
-
lm-human-preference-details
-
-
-
-
-
-
-
-
-
-
-
-
-
-
512
0.2
0.2
0.00001
1
1
0.95
64
64
64
0.00001
512
1
4
1953
1000000
0.1
true
8
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
19m 14s
-
gpt2-xl
32
true
0
false
-
0.00001
train_reward_accelerate_pt_adam_gpt2_xl_grad_accu
-
1
-
sentiment/offline_5k.json
4
4992
-
-
4
4
256
-
512
0.00005
true
true
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
19m 12s
-
gpt2-xl
32
true
0
false
-
0.00001
train_reward_accelerate_pt_adam_gpt2_xl_grad_accu
-
1
-
sentiment/offline_5k.json
4
4992
-
-
4
4
256
-
512
0.00005
true
true
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
19m 59s
-
gpt2-xl
32
true
0
false
-
0.00001
train_reward_accelerate_pt_adam_gpt2_xl_grad_accu
-
1
-
sentiment/offline_5k.json
4
4992
-
-
4
4
256
-
512
0.00005
true
true
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
19m 27s
-
gpt2-xl
32
true
0
false
-
0.00001
train_reward_accelerate_pt_adam_gpt2_xl_grad_accu
-
1
-
sentiment/offline_5k.json
4
4992
-
-
4
4
256
-
512
0.00005
true
true
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Finished
costa-huang
1.0.0b1
gpt2-xl
pr-24
pt_adam
19m 29s
-
gpt2-xl
32
true
0
false
-
0.00001
train_reward_accelerate_pt_adam_gpt2_xl_grad_accu
-
1
-
sentiment/offline_5k.json
4
4992
-
-
4
4
256
-
512
0.00005
true
true
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1-20
of 1,067