Akhauriyash's workspace
Runs
7
State
Notes
User
Tags
Created
Runtime
Sweep
checkpoint.continue_training_from_init
checkpoint.dump.every
checkpoint.dump.keep
checkpoint.eval.every
checkpoint.eval.keep
checkpoint.path
data.add_bos
data.add_eos
data.batch_size
data.load_async
data.n_views
data.prefetch_size
data.root_dir
data.seed
data.seq_len
data.sources.dclm_baseline_1.0_shuffled
data.tokenizer.name
data.tokenizer.path
distributed.compile
distributed.compile_cache_size_limit
distributed.detect_anomaly
distributed.dp_replicate
distributed.dp_shard
distributed.float8_filter
distributed.fsdp_type
distributed.matmul_allow_tf32
distributed.model_dtype
distributed.selective_activation_checkpointing
distributed.spawn_method
distributed.tp_size
dump_dir
env.ENABLE_INTRA_NODE_COMM
env.MKL_NUM_THREADS
env.MKL_SERVICE_FORCE_INTEL
env.NCCL_DEBUG
env.NCCL_IB_TIMEOUT
env.OMP_NUM_THREADS
env.TORCH_NCCL_ASYNC_ERROR_HANDLING
env.TORCH_NCCL_AVOID_RECORD_STREAMS
eval.generator.dtype
eval.generator.max_tokens
eval.harness.tasks
gc_collect_freq
grad_acc_steps
Finished
-
akhauriyash
2h 33m 34s
-
false
10000
3
5000
-1
/scratch/ya255/lingua/comparative_runs/isoFLOP104/checkpoints
true
true
16
true
2
512
/scratch/ya255/lingua/setup/data/
42
1024
100
sp
setup/l2tokenizer/tokenizer.model
false
8
false
1
1
layers\.[0-9]+\.
full_shard
false
bf16
false
forkserver
1
/scratch/ya255/lingua/comparative_runs/isoFLOP104
1
1
GNU
INFO
22
1
1
1
bf16
1024
["wikitext"]
1000
1
Finished
-
akhauriyash
4h 50m 39s
-
false
10000
3
5000
-1
/scratch/ya255/lingua/comparative_runs/isoAttentiveSSM8/checkpoints
true
true
16
true
2
512
/scratch/ya255/lingua/setup/data/
42
1024
100
sp
setup/l2tokenizer/tokenizer.model
false
8
false
1
1
layers\.[0-9]+\.
full_shard
false
bf16
false
forkserver
1
/scratch/ya255/lingua/comparative_runs/isoAttentiveSSM8
1
1
GNU
INFO
22
1
1
1
bf16
1024
["wikitext"]
1000
1
Finished
-
akhauriyash
1h 26m 58s
-
false
10000
3
5000
-1
/scratch/ya255/lingua/comparative_runs/isoKV4/checkpoints
true
true
16
true
2
512
/scratch/ya255/lingua/setup/data/
42
1024
100
sp
setup/l2tokenizer/tokenizer.model
true
8
false
1
1
layers\.[0-9]+\.
full_shard
false
bf16
false
forkserver
1
/scratch/ya255/lingua/comparative_runs/isoKV4
1
1
GNU
INFO
22
1
1
1
bf16
1024
["wikitext"]
1000
1
Finished
-
akhauriyash
1h 20m 28s
-
false
10000
3
5000
-1
/scratch/ya255/lingua/comparative_runs/isoKV8/checkpoints
true
true
16
true
2
512
/scratch/ya255/lingua/setup/data/
42
1024
100
sp
setup/l2tokenizer/tokenizer.model
true
8
false
1
1
layers\.[0-9]+\.
full_shard
false
bf16
false
forkserver
1
/scratch/ya255/lingua/comparative_runs/isoKV8
1
1
GNU
INFO
22
1
1
1
bf16
1024
["wikitext"]
1000
1
Finished
-
akhauriyash
2h 38m 7s
-
false
10000
3
5000
-1
/scratch/ya255/lingua/comparative_runs/isoFLOP160/checkpoints
true
true
16
true
2
512
/scratch/ya255/lingua/setup/data/
42
1024
100
sp
setup/l2tokenizer/tokenizer.model
false
8
false
1
1
layers\.[0-9]+\.
full_shard
false
bf16
false
forkserver
1
/scratch/ya255/lingua/comparative_runs/isoFLOP160
1
1
GNU
INFO
22
1
1
1
bf16
1024
["wikitext"]
1000
1
Finished
-
akhauriyash
4h 50m 45s
-
false
10000
3
5000
-1
/scratch/ya255/lingua/comparative_runs/isoAttentiveSSM4/checkpoints
true
true
16
true
2
512
/scratch/ya255/lingua/setup/data/
42
1024
100
sp
setup/l2tokenizer/tokenizer.model
false
8
false
1
1
layers\.[0-9]+\.
full_shard
false
bf16
false
forkserver
1
/scratch/ya255/lingua/comparative_runs/isoAttentiveSSM4
1
1
GNU
INFO
22
1
1
1
bf16
1024
["wikitext"]
1000
1
Finished
-
akhauriyash
1h 38m 44s
-
false
10000
3
5000
-1
/scratch/ya255/lingua/comparative_runs/isoParam/checkpoints
true
true
16
true
2
512
/scratch/ya255/lingua/setup/data/
42
1024
100
sp
setup/l2tokenizer/tokenizer.model
true
8
false
1
1
layers\.[0-9]+\.
full_shard
false
bf16
false
forkserver
1
/scratch/ya255/lingua/comparative_runs/isoParam
1
1
GNU
INFO
22
1
1
1
bf16
1024
["wikitext"]
1000
1
1-7
of 7