598 lines
397 KiB
Plaintext
598 lines
397 KiB
Plaintext
./slurm_setup.sh: line 48: GLOBAL_TRAIN_BATCH_SIZE / NNODES / GPUS_PER_NODE / GRADIENT_ACCUMULATION_STEPS: division by 0 (error token is "GRADIENT_ACCUMULATION_STEPS")
|
|
./slurm_setup.sh: line 48: GLOBAL_TRAIN_BATCH_SIZE / NNODES / GPUS_PER_NODE / GRADIENT_ACCUMULATION_STEPS: division by 0 (error token is "GRADIENT_ACCUMULATION_STEPS")
|
|
./slurm_setup.sh: line 48: GLOBAL_TRAIN_BATCH_SIZE / NNODES / GPUS_PER_NODE / GRADIENT_ACCUMULATION_STEPS: division by 0 (error token is "GRADIENT_ACCUMULATION_STEPS")
|
|
./slurm_setup.sh: line 48: GLOBAL_TRAIN_BATCH_SIZE / NNODES / GPUS_PER_NODE / GRADIENT_ACCUMULATION_STEPS: division by 0 (error token is "GRADIENT_ACCUMULATION_STEPS")
|
|
./slurm_setup.sh: line 48: GLOBAL_TRAIN_BATCH_SIZE / NNODES / GPUS_PER_NODE / GRADIENT_ACCUMULATION_STEPS: division by 0 (error token is "GRADIENT_ACCUMULATION_STEPS")
|
|
./slurm_setup.sh: line 48: GLOBAL_TRAIN_BATCH_SIZE / NNODES / GPUS_PER_NODE / GRADIENT_ACCUMULATION_STEPS: division by 0 (error token is "GRADIENT_ACCUMULATION_STEPS")
|
|
./slurm_setup.sh: line 48: GLOBAL_TRAIN_BATCH_SIZE / NNODES / GPUS_PER_NODE / GRADIENT_ACCUMULATION_STEPS: division by 0 (error token is "GRADIENT_ACCUMULATION_STEPS")
|
|
./slurm_setup.sh: line 48: GLOBAL_TRAIN_BATCH_SIZE / NNODES / GPUS_PER_NODE / GRADIENT_ACCUMULATION_STEPS: division by 0 (error token is "GRADIENT_ACCUMULATION_STEPS")
|
|
2026-04-13 22:52:20,919 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:20,919 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:20,919 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:20,919 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:20,919 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:20,919 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:20,919 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:20,919 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:20,919 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:20,919 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:20,919 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:20,919 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:20,919 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:20,919 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:20,919 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:20,919 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,019 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,019 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,019 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,019 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,019 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,019 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,019 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,019 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,019 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,019 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,019 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,019 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,019 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,019 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,019 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,019 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
2026-04-13 22:52:21,054 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,054 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,054 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,054 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,054 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,054 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,054 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,054 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,054 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,054 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,054 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,054 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,054 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,054 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,054 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,054 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,147 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,147 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,147 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,147 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,147 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,147 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,147 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,147 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,147 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,147 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,147 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,147 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,147 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,147 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,147 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,147 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,153 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,153 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,153 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,153 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,153 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,153 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,153 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,153 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,153 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,153 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,153 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,153 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,153 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,153 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,153 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,154 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
2026-04-13 22:52:21,229 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,229 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,229 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,229 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,229 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,229 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,229 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,229 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,229 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,229 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,229 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,229 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,229 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,229 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,229 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,229 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
2026-04-13 22:52:21,255 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,255 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,255 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,255 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,255 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,255 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,255 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,255 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,255 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,255 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,255 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,255 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,255 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,255 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,255 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,255 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
2026-04-13 22:52:21,396 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,396 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,396 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,396 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,396 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,396 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,396 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,396 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,396 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,396 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,396 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,396 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,396 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,396 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
2026-04-13 22:52:21,396 - INFO - Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
|
|
2026-04-13 22:52:21,396 - INFO - Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
/home/ligengz/anaconda3/lib/python3.11/site-packages/pandas/core/arrays/masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).
|
|
from pandas.core import (
|
|
2026-04-13 22:52:22,026 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,026 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,026 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,026 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,026 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,026 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,026 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,026 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,184 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,186 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,186 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,186 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,186 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,186 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,186 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,187 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,216 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,216 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,216 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,216 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,217 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,216 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,217 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,217 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,227 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,227 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,227 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,227 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,227 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,227 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,227 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,227 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,288 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,288 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,288 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,288 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,288 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,288 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,288 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,289 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,333 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,333 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,333 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,333 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,333 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,333 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,333 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,333 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,495 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,496 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,496 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,496 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,496 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,496 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,496 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:22,497 - INFO - PyTorch version 2.6.0 available.
|
|
2026-04-13 22:52:59,446 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-57_batch-block1-3544', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,542 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-1062', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 3, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,543 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-1062', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 6, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,544 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3273', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 2, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,559 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-1062', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 7, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,561 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_2026-04-13 22:52:59,561 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3273', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 7, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-57_batch-block1-3544', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 5, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,580 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-57_batch-block1-3544', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 6, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,581 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-57_batch-block1-3544', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 3, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,587 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3035', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 4, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,588 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3035', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 5, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,599 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-57_batch-block1-3544', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 7, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,600 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-57_batch-block1-3544', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 4, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,603 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3035', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 6, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,605 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-57_batch-block1-3544', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 2, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules'2026-04-13 22:52:59,605 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_: None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3035', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 3, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,606 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3035', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 1, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,842 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3578', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:52:59,905 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3339', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,103 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3578', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 4, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,103 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3578', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 1, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,105 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3578', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 3, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,105 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3578', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 2, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,105 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3578', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 7, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,105 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3339', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 1, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,108 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3339', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 7, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,120 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3339', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 6, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,126 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_2026-04-13 22:53:00,126 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3339', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 2, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3578', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 5, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,126 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3578', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 6, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,127 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3339', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 4, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,128 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3339', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 3, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:00,130 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-59_batch-block1-3339', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 5, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,187 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-1089', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,192 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-3736', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,393 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-1089', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 7, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,393 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-1089', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 5, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,407 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-1089', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 6, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,411 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-1089', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 3, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,428 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-1089', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 2, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,428 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-1089', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 4, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,432 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-1089', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 1, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,437 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-3736', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 2, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,457 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-3736', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 5, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,458 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-3736', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 1, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,459 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-3736', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 7, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,460 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-3736', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 3, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,475 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-3736', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 4, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:02,476 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-53-00_batch-block1-3736', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 6, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,530 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-57_batch-block1-3544', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 1, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,546 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3035', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,573 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-1062', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,603 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3273', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,646 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3035', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 7, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,647 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3035', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 2, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,668 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-1062', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 4, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,687 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-1062', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 1, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,687 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-1062', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 5, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,701 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-1062', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 2, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,717 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3273', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 5, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,736 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3273', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 6, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,741 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3273', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 4, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,743 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3273', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 3, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
2026-04-13 22:53:05,744 - INFO - Training config: {'model_name': 'runs/dev/tw-8node-resume/checkpoint-64', 'template_name': 'qwen', 'block_size': 40960, 'wandb_project': 'ThreadWeaver', 'train_file_path': './data-04-13', 'dagger': False, 'attn_implementation': 'flex_attention', 'output_dir': 'runs/dev/tw-8node-resume', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': <IntervalStrategy.NO: 'no'>, 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0001, 'adam_beta1': 0.9, 'adam_beta2': 0.95, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 8.0, 'max_steps': -1, 'lr_scheduler_type': <SchedulerType.COSINE: 'cosine'>, 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.05, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'runs/dev/tw-8node-resume/runs/Apr13_22-52-58_batch-block1-3273', 'logging_strategy': <IntervalStrategy.STEPS: 'steps'>, 'logging_first_step': False, 'logging_steps': 1.0, 'logging_nan_inf_filter': True, 'save_strategy': <SaveStrategy.STEPS: 'steps'>, 'save_steps': 20, 'save_total_limit': 2, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 1, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'runs/dev/tw-8node-resume', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'tp_size': 0, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, 'deepspeed': 'configs/deepspeed_zero3.json', 'label_smoothing_factor': 0.0, 'optim': <OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': 'runs/dev/tw-8node-resume/checkpoint-64', 'hub_model_id': None, 'hub_strategy': <HubStrategy.EVERY_SAVE: 'every_save'>, 'hub_token': None, 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': None, '_n_gpu': 1, 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': True, 'eval_use_gather_object': False, 'average_tokens_across_devices': True, 'model_init_kwargs': None, 'chat_template_path': None, 'dataset_text_field': 'qwen_text', 'dataset_kwargs': None, 'dataset_num_proc': None, 'eos_token': None, 'pad_token': None, 'max_length': 1024, 'packing': False, 'packing_strategy': 'ffd', 'padding_free': False, 'pad_to_multiple_of': None, 'eval_packing': None, 'completion_only_loss': None, 'assistant_only_loss': False, 'activation_offloading': False, 'max_seq_length': None}
|
|
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 2
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 2
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 2
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.80s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.80s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.80s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.80s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.80s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.80s/it]
|
|
Loading checkpoint shards: 2
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 2
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 2
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 2
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████5%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████5%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████5%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.87s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████5%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.84s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████5%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.05s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████5%|██▌ | 1/4 [00:04<00:14, 4.80s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:04<00:14, 4.80s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.03s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.03s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.03s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.03s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.03s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.03s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.03s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.03s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████5%|██▌ | 1/4 [00:04<00:14, 4.86s/it]
|
|
Loading checkpoint shards: 25%|██▌ | 1/4 [00:05<00:15, 5.25s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.06s/it]
|
|
Loading checkpoint shards: 50%|█████ | 2/4 [00:08<00:08, 4.16s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.63s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
00, 2.89s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.89s/it]
|
|
▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.64s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.89s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.89s/it]
|
|
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.89s/it]
|
|
00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.89s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.89s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.89s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.90s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.25s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:11<00:00, 2.91s/it]
|
|
|
|
Loading checkpoint shards: 75%|███████▌ | 3/4 [00:11<00:03, 3.69s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:12<00:00, 2.62s/it]
|
|
Loading checkpoint shards: 100%|██████████| 4/4 [00:12<00:00, 3.19s/it]
|
|
|
|
Tokenizing train dataset: 0%| | 0/964 [00:00<?, ? examples/s]
|
|
Tokenizing train dataset: 0%| | 4/964 [00:00<00:33, 28.97 examples/s]
|
|
Tokenizing train dataset: 1%| | 8/964 [00:00<00:29, 32.55 examples/s]
|
|
Tokenizing train dataset: 1%| | 12/964 [00:00<00:29, 32.36 examples/s]
|
|
Tokenizing train dataset: 2%|▏ | 16/964 [00:00<00:31, 30.35 examples/s]
|
|
Tokenizing train dataset: 2%|▏ | 21/964 [00:00<00:26, 35.75 examples/s]
|
|
Tokenizing train dataset: 3%|▎ | 25/964 [00:00<00:26, 35.68 examples/s]
|
|
Tokenizing train dataset: 3%|▎ | 29/964 [00:00<00:28, 33.08 examples/s]
|
|
Tokenizing train dataset: 4%|▎ | 34/964 [00:01<00:31, 29.36 examples/s]
|
|
Tokenizing train dataset: 4%|▍ | 38/964 [00:01<00:32, 28.24 examples/s]
|
|
Tokenizing train dataset: 4%|▍ | 42/964 [00:01<00:36, 25.50 examples/s]
|
|
Tokenizing train dataset: 5%|▍ | 48/964 [00:01<00:30, 29.97 examples/s]
|
|
Tokenizing train dataset: 5%|▌ | 53/964 [00:01<00:31, 29.14 examples/s]
|
|
Tokenizing train dataset: 6%|▌ | 58/964 [00:01<00:35, 25.77 examples/s]
|
|
Tokenizing train dataset: 6%|▋ | 62/964 [00:02<00:33, 27.00 examples/s]
|
|
Tokenizing train dataset: 7%|▋ | 66/964 [00:02<00:32, 27.31 examples/s]
|
|
Tokenizing train dataset: 7%|▋ | 70/964 [00:02<00:31, 28.73 examples/s]
|
|
Tokenizing train dataset: 8%|▊ | 75/964 [00:02<00:28, 31.43 examples/s]
|
|
Tokenizing train dataset: 8%|▊ | 80/964 [00:02<00:25, 34.01 examples/s]
|
|
Tokenizing train dataset: 9%|▉ | 85/964 [00:02<00:26, 33.22 examples/s]
|
|
Tokenizing train dataset: 9%|▉ | 89/964 [00:02<00:27, 31.55 examples/s]
|
|
Tokenizing train dataset: 10%|▉ | 95/964 [00:03<00:26, 32.53 examples/s]
|
|
Tokenizing train dataset: 10%|█ | 99/964 [00:03<00:27, 31.38 examples/s]
|
|
Tokenizing train dataset: 11%|█ | 104/964 [00:03<00:32, 26.61 examples/s]
|
|
Tokenizing train dataset: 11%|█ | 108/964 [00:03<00:31, 27.34 examples/s]
|
|
Tokenizing train dataset: 12%|█▏ | 113/964 [00:03<00:29, 28.74 examples/s]
|
|
Tokenizing train dataset: 12%|█▏ | 116/964 [00:03<00:30, 27.62 examples/s]
|
|
Tokenizing train dataset: 13%|█▎ | 121/964 [00:04<00:29, 28.80 examples/s]
|
|
Tokenizing train dataset: 13%|█▎ | 126/964 [00:04<00:27, 30.66 examples/s]
|
|
Tokenizing train dataset: 14%|█▎ | 132/964 [00:04<00:27, 29.81 examples/s]
|
|
Tokenizing train dataset: 14%|█▍ | 137/964 [00:04<00:30, 26.92 examples/s]
|
|
Tokenizing train dataset: 15%|█▍ | 141/964 [00:04<00:32, 25.72 examples/s]
|
|
Tokenizing train dataset: 15%|█▌ | 145/964 [00:04<00:29, 27.80 examples/s]
|
|
Tokenizing train dataset: 15%|█▌ | 149/964 [00:05<00:28, 28.79 examples/s]
|
|
Tokenizing train dataset: 16%|█▌ | 153/964 [00:05<00:26, 30.30 examples/s]
|
|
Tokenizing train dataset: 16%|█▋ | 158/964 [00:05<00:25, 31.82 examples/s]
|
|
Tokenizing train dataset: 17%|█▋ | 163/964 [00:05<00:27, 28.80 examples/s]
|
|
Tokenizing train dataset: 17%|█▋ | 168/964 [00:05<00:26, 30.19 examples/s]
|
|
Tokenizing train dataset: 18%|█▊ | 172/964 [00:05<00:28, 27.91 examples/s]
|
|
Tokenizing train dataset: 18%|█▊ | 176/964 [00:05<00:27, 28.71 examples/s]
|
|
Tokenizing train dataset: 19%|█▊ | 180/964 [00:06<00:26, 30.09 examples/s]
|
|
Tokenizing train dataset: 19%|█▉ | 184/964 [00:06<00:33, 23.35 examples/s]
|
|
Tokenizing train dataset: 19%|█▉ | 187/964 [00:06<00:33, 23.13 examples/s]
|
|
Tokenizing train dataset: 20%|█▉ | 190/964 [00:06<00:35, 22.09 examples/s]
|
|
Tokenizing train dataset: 20%|██ | 193/964 [00:06<00:37, 20.49 examples/s]
|
|
Tokenizing train dataset: 20%|██ | 196/964 [00:06<00:34, 21.98 examples/s]
|
|
Tokenizing train dataset: 21%|██ | 201/964 [00:07<00:31, 24.34 examples/s]
|
|
Tokenizing train dataset: 21%|██▏ | 205/964 [00:07<00:31, 24.35 examples/s]
|
|
Tokenizing train dataset: 22%|██▏ | 209/964 [00:07<00:30, 24.37 examples/s]
|
|
Tokenizing train dataset: 22%|██▏ | 213/964 [00:07<00:29, 25.06 examples/s]
|
|
Tokenizing train dataset: 22%|██▏ | 216/964 [00:07<00:29, 24.94 examples/s]
|
|
Tokenizing train dataset: 23%|██▎ | 221/964 [00:07<00:25, 29.00 examples/s]
|
|
Tokenizing train dataset: 23%|██▎ | 224/964 [00:07<00:27, 27.15 examples/s]
|
|
Tokenizing train dataset: 24%|██▍ | 229/964 [00:08<00:25, 28.42 examples/s]
|
|
Tokenizing train dataset: 24%|██▍ | 232/964 [00:08<00:26, 28.14 examples/s]
|
|
Tokenizing train dataset: 24%|██▍ | 235/964 [00:08<00:28, 25.65 examples/s]
|
|
Tokenizing train dataset: 25%|██▍ | 239/964 [00:08<00:29, 24.67 examples/s]
|
|
Tokenizing train dataset: 25%|██▌ | 243/964 [00:08<00:29, 24.65 examples/s]
|
|
Tokenizing train dataset: 26%|██▌ | 248/964 [00:08<00:24, 28.80 examples/s]
|
|
Tokenizing train dataset: 26%|██▌ | 252/964 [00:08<00:23, 30.22 examples/s]
|
|
Tokenizing train dataset: 27%|██▋ | 256/964 [00:09<00:22, 31.54 examples/s]
|
|
Tokenizing train dataset: 27%|██▋ | 260/964 [00:09<00:24, 29.16 examples/s]
|
|
Tokenizing train dataset: 27%|██▋ | 264/964 [00:09<00:23, 29.88 examples/s]
|
|
Tokenizing train dataset: 28%|██▊ | 269/964 [00:09<00:21, 32.77 examples/s]
|
|
Tokenizing train dataset: 28%|██▊ | 274/964 [00:09<00:25, 26.70 examples/s]
|
|
Tokenizing train dataset: 29%|██▊ | 277/964 [00:09<00:26, 25.55 examples/s]
|
|
Tokenizing train dataset: 29%|██▉ | 282/964 [00:10<00:23, 28.94 examples/s]
|
|
Tokenizing train dataset: 30%|██▉ | 287/964 [00:10<00:23, 29.43 examples/s]
|
|
Tokenizing train dataset: 30%|███ | 291/964 [00:10<00:23, 28.65 examples/s]
|
|
Tokenizing train dataset: 31%|███ | 295/964 [00:10<00:23, 28.82 examples/s]
|
|
Tokenizing train dataset: 31%|███ | 298/964 [00:10<00:23, 27.97 examples/s]
|
|
Tokenizing train dataset: 31%|███ | 301/964 [00:10<00:24, 26.55 examples/s]
|
|
Tokenizing train dataset: 32%|███▏ | 305/964 [00:10<00:24, 27.24 examples/s]
|
|
Tokenizing train dataset: 32%|███▏ | 309/964 [00:10<00:22, 28.66 examples/s]
|
|
Tokenizing train dataset: 32%|███▏ | 312/964 [00:11<00:23, 28.31 examples/s]
|
|
Tokenizing train dataset: 33%|███▎ | 316/964 [00:11<00:21, 29.87 examples/s]
|
|
Tokenizing train dataset: 33%|███▎ | 322/964 [00:11<00:19, 33.63 examples/s]
|
|
Tokenizing train dataset: 34%|███▍ | 326/964 [00:11<00:20, 31.39 examples/s]
|
|
Tokenizing train dataset: 34%|███▍ | 331/964 [00:11<00:18, 34.47 examples/s]
|
|
Tokenizing train dataset: 35%|███▍ | 335/964 [00:11<00:19, 32.30 examples/s]
|
|
Tokenizing train dataset: 35%|███▌ | 339/964 [00:11<00:20, 29.80 examples/s]
|
|
Tokenizing train dataset: 36%|███▌ | 343/964 [00:12<00:22, 27.76 examples/s]
|
|
Tokenizing train dataset: 36%|███▌ | 347/964 [00:12<00:21, 29.12 examples/s]
|
|
Tokenizing train dataset: 37%|███▋ | 353/964 [00:12<00:18, 33.37 examples/s]
|
|
Tokenizing train dataset: 37%|███▋ | 357/964 [00:12<00:18, 33.54 examples/s]
|
|
Tokenizing train dataset: 38%|███▊ | 363/964 [00:12<00:17, 35.16 examples/s]
|
|
Tokenizing train dataset: 38%|███▊ | 367/964 [00:12<00:18, 32.52 examples/s]
|
|
Tokenizing train dataset: 39%|███▊ | 372/964 [00:12<00:19, 30.92 examples/s]
|
|
Tokenizing train dataset: 39%|███▉ | 377/964 [00:13<00:19, 29.71 examples/s]
|
|
Tokenizing train dataset: 40%|███▉ | 381/964 [00:13<00:20, 28.49 examples/s]
|
|
Tokenizing train dataset: 40%|████ | 386/964 [00:13<00:19, 30.07 examples/s]
|
|
Tokenizing train dataset: 40%|████ | 390/964 [00:13<00:19, 28.77 examples/s]
|
|
Tokenizing train dataset: 41%|████ | 394/964 [00:13<00:19, 29.11 examples/s]
|
|
Tokenizing train dataset: 41%|████▏ | 399/964 [00:13<00:20, 28.21 examples/s]
|
|
Tokenizing train dataset: 42%|████▏ | 402/964 [00:14<00:21, 26.03 examples/s]
|
|
Tokenizing train dataset: 42%|████▏ | 407/964 [00:14<00:19, 28.67 examples/s]
|
|
Tokenizing train dataset: 43%|████▎ | 411/964 [00:14<00:20, 27.09 examples/s]
|
|
Tokenizing train dataset: 43%|████▎ | 415/964 [00:14<00:20, 27.04 examples/s]
|
|
Tokenizing train dataset: 43%|████▎ | 419/964 [00:14<00:20, 27.13 examples/s]
|
|
Tokenizing train dataset: 44%|████▍ | 423/964 [00:14<00:20, 26.81 examples/s]
|
|
Tokenizing train dataset: 44%|████▍ | 426/964 [00:14<00:20, 26.69 examples/s]
|
|
Tokenizing train dataset: 45%|████▍ | 431/964 [00:15<00:17, 30.02 examples/s]
|
|
Tokenizing train dataset: 45%|████▌ | 435/964 [00:15<00:18, 28.05 examples/s]
|
|
Tokenizing train dataset: 46%|████▌ | 439/964 [00:15<00:18, 28.48 examples/s]
|
|
Tokenizing train dataset: 46%|████▌ | 445/964 [00:15<00:16, 31.56 examples/s]
|
|
Tokenizing train dataset: 47%|████▋ | 451/964 [00:15<00:16, 30.23 examples/s]
|
|
Tokenizing train dataset: 47%|████▋ | 455/964 [00:15<00:17, 29.88 examples/s]
|
|
Tokenizing train dataset: 48%|████▊ | 460/964 [00:16<00:17, 28.36 examples/s]
|
|
Tokenizing train dataset: 48%|████▊ | 464/964 [00:16<00:17, 29.05 examples/s]
|
|
Tokenizing train dataset: 49%|████▊ | 469/964 [00:16<00:17, 28.79 examples/s]
|
|
Tokenizing train dataset: 49%|████▉ | 473/964 [00:16<00:16, 29.87 examples/s]
|
|
Tokenizing train dataset: 50%|████▉ | 479/964 [00:16<00:15, 31.99 examples/s]
|
|
Tokenizing train dataset: 50%|█████ | 483/964 [00:16<00:16, 29.05 examples/s]
|
|
Tokenizing train dataset: 50%|█████ | 486/964 [00:17<00:18, 25.60 examples/s]
|
|
Tokenizing train dataset: 51%|█████ | 491/964 [00:17<00:16, 27.89 examples/s]
|
|
Tokenizing train dataset: 51%|█████▏ | 496/964 [00:17<00:16, 27.85 examples/s]
|
|
Tokenizing train dataset: 52%|█████▏ | 499/964 [00:17<00:16, 27.69 examples/s]
|
|
Tokenizing train dataset: 52%|█████▏ | 503/964 [00:17<00:16, 28.32 examples/s]
|
|
Tokenizing train dataset: 52%|█████▏ | 506/964 [00:17<00:17, 25.93 examples/s]
|
|
Tokenizing train dataset: 53%|█████▎ | 510/964 [00:17<00:16, 27.99 examples/s]
|
|
Tokenizing train dataset: 53%|█████▎ | 513/964 [00:17<00:16, 27.18 examples/s]
|
|
Tokenizing train dataset: 54%|█████▎ | 517/964 [00:18<00:17, 26.03 examples/s]
|
|
Tokenizing train dataset: 54%|█████▍ | 521/964 [00:18<00:16, 26.27 examples/s]
|
|
Tokenizing train dataset: 55%|█████▍ | 527/964 [00:18<00:14, 30.67 examples/s]
|
|
Tokenizing train dataset: 55%|█████▌ | 531/964 [00:18<00:13, 32.36 examples/s]
|
|
Tokenizing train dataset: 56%|█████▌ | 536/964 [00:18<00:13, 30.99 examples/s]
|
|
Tokenizing train dataset: 56%|█████▌ | 541/964 [00:18<00:13, 31.48 examples/s]
|
|
Tokenizing train dataset: 57%|█████▋ | 545/964 [00:19<00:14, 29.49 examples/s]
|
|
Tokenizing train dataset: 57%|█████▋ | 550/964 [00:19<00:13, 31.32 examples/s]
|
|
Tokenizing train dataset: 57%|█████▋ | 554/964 [00:19<00:16, 25.36 examples/s]
|
|
Tokenizing train dataset: 58%|█████▊ | 559/964 [00:19<00:15, 25.65 examples/s]
|
|
Tokenizing train dataset: 58%|█████▊ | 562/964 [00:19<00:16, 24.40 examples/s]
|
|
Tokenizing train dataset: 59%|█████▊ | 565/964 [00:19<00:16, 24.59 examples/s]
|
|
Tokenizing train dataset: 59%|█████▉ | 569/964 [00:19<00:15, 25.81 examples/s]
|
|
Tokenizing train dataset: 60%|█████▉ | 574/964 [00:20<00:13, 27.98 examples/s]
|
|
Tokenizing train dataset: 60%|██████ | 579/964 [00:20<00:12, 30.33 examples/s]
|
|
Tokenizing train dataset: 61%|██████ | 584/964 [00:20<00:13, 29.16 examples/s]
|
|
Tokenizing train dataset: 61%|██████ | 589/964 [00:20<00:11, 32.95 examples/s]
|
|
Tokenizing train dataset: 62%|██████▏ | 593/964 [00:20<00:10, 34.00 examples/s]
|
|
Tokenizing train dataset: 62%|██████▏ | 598/964 [00:20<00:11, 31.61 examples/s]
|
|
Tokenizing train dataset: 63%|██████▎ | 603/964 [00:21<00:11, 30.12 examples/s]
|
|
Tokenizing train dataset: 63%|██████▎ | 609/964 [00:21<00:11, 31.24 examples/s]
|
|
Tokenizing train dataset: 64%|██████▎ | 613/964 [00:21<00:11, 30.87 examples/s]
|
|
Tokenizing train dataset: 64%|██████▍ | 618/964 [00:21<00:11, 30.08 examples/s]
|
|
Tokenizing train dataset: 65%|██████▍ | 622/964 [00:21<00:11, 29.70 examples/s]
|
|
Tokenizing train dataset: 65%|██████▌ | 627/964 [00:21<00:10, 30.82 examples/s]
|
|
Tokenizing train dataset: 66%|██████▌ | 632/964 [00:22<00:12, 26.36 examples/s]
|
|
Tokenizing train dataset: 66%|██████▌ | 635/964 [00:22<00:12, 25.34 examples/s]
|
|
Tokenizing train dataset: 66%|██████▋ | 639/964 [00:22<00:11, 27.17 examples/s]
|
|
Tokenizing train dataset: 67%|██████▋ | 643/964 [00:22<00:11, 27.10 examples/s]
|
|
Tokenizing train dataset: 67%|██████▋ | 648/964 [00:22<00:11, 28.46 examples/s]
|
|
Tokenizing train dataset: 68%|██████▊ | 652/964 [00:22<00:10, 30.39 examples/s]
|
|
Tokenizing train dataset: 68%|██████▊ | 656/964 [00:22<00:10, 29.08 examples/s]
|
|
Tokenizing train dataset: 69%|██████▊ | 661/964 [00:23<00:09, 30.48 examples/s]
|
|
Tokenizing train dataset: 69%|██████▉ | 665/964 [00:23<00:10, 29.35 examples/s]
|
|
Tokenizing train dataset: 69%|██████▉ | 668/964 [00:23<00:10, 28.62 examples/s]
|
|
Tokenizing train dataset: 70%|██████▉ | 673/964 [00:23<00:09, 29.42 examples/s]
|
|
Tokenizing train dataset: 70%|███████ | 676/964 [00:23<00:10, 28.63 examples/s]
|
|
Tokenizing train dataset: 71%|███████ | 680/964 [00:23<00:09, 28.49 examples/s]
|
|
Tokenizing train dataset: 71%|███████ | 685/964 [00:23<00:09, 30.77 examples/s]
|
|
Tokenizing train dataset: 72%|███████▏ | 690/964 [00:24<00:10, 26.41 examples/s]
|
|
Tokenizing train dataset: 72%|███████▏ | 694/964 [00:24<00:09, 27.45 examples/s]
|
|
Tokenizing train dataset: 73%|███████▎ | 699/964 [00:24<00:08, 30.36 examples/s]
|
|
Tokenizing train dataset: 73%|███████▎ | 704/964 [00:24<00:08, 30.25 examples/s]
|
|
Tokenizing train dataset: 74%|███████▎ | 710/964 [00:24<00:07, 34.10 examples/s]
|
|
Tokenizing train dataset: 74%|███████▍ | 714/964 [00:24<00:07, 33.94 examples/s]
|
|
Tokenizing train dataset: 75%|███████▍ | 719/964 [00:24<00:06, 35.10 examples/s]
|
|
Tokenizing train dataset: 75%|███████▌ | 723/964 [00:25<00:07, 30.75 examples/s]
|
|
Tokenizing train dataset: 76%|███████▌ | 728/964 [00:25<00:07, 31.43 examples/s]
|
|
Tokenizing train dataset: 76%|███████▌ | 734/964 [00:25<00:07, 30.26 examples/s]
|
|
Tokenizing train dataset: 77%|███████▋ | 738/964 [00:25<00:07, 29.29 examples/s]
|
|
Tokenizing train dataset: 77%|███████▋ | 742/964 [00:25<00:08, 24.79 examples/s]
|
|
Tokenizing train dataset: 77%|███████▋ | 746/964 [00:26<00:08, 24.36 examples/s]
|
|
Tokenizing train dataset: 78%|███████▊ | 750/964 [00:26<00:08, 25.85 examples/s]
|
|
Tokenizing train dataset: 78%|███████▊ | 753/964 [00:26<00:07, 26.55 examples/s]
|
|
Tokenizing train dataset: 79%|███████▊ | 758/964 [00:26<00:07, 28.89 examples/s]
|
|
Tokenizing train dataset: 79%|███████▉ | 763/964 [00:26<00:06, 31.56 examples/s]
|
|
Tokenizing train dataset: 80%|███████▉ | 769/964 [00:26<00:05, 35.66 examples/s]
|
|
Tokenizing train dataset: 80%|████████ | 774/964 [00:26<00:05, 33.17 examples/s]
|
|
Tokenizing train dataset: 81%|████████ | 780/964 [00:27<00:06, 29.89 examples/s]
|
|
Tokenizing train dataset: 81%|████████▏ | 785/964 [00:27<00:05, 31.08 examples/s]
|
|
Tokenizing train dataset: 82%|████████▏ | 790/964 [00:27<00:05, 32.55 examples/s]
|
|
Tokenizing train dataset: 82%|████████▏ | 794/964 [00:27<00:05, 32.97 examples/s]
|
|
Tokenizing train dataset: 83%|████████▎ | 798/964 [00:27<00:05, 32.63 examples/s]
|
|
Tokenizing train dataset: 83%|████████▎ | 802/964 [00:27<00:05, 32.02 examples/s]
|
|
Tokenizing train dataset: 84%|████████▍ | 808/964 [00:27<00:04, 31.74 examples/s]
|
|
Tokenizing train dataset: 84%|████████▍ | 812/964 [00:28<00:05, 28.92 examples/s]
|
|
Tokenizing train dataset: 85%|████████▍ | 816/964 [00:28<00:05, 27.45 examples/s]
|
|
Tokenizing train dataset: 85%|████████▍ | 819/964 [00:28<00:05, 27.05 examples/s]
|
|
Tokenizing train dataset: 85%|████████▌ | 824/964 [00:28<00:05, 27.42 examples/s]
|
|
Tokenizing train dataset: 86%|████████▌ | 827/964 [00:28<00:05, 27.27 examples/s]
|
|
Tokenizing train dataset: 86%|████████▌ | 831/964 [00:28<00:05, 25.13 examples/s]
|
|
Tokenizing train dataset: 87%|████████▋ | 835/964 [00:28<00:04, 26.76 examples/s]
|
|
Tokenizing train dataset: 87%|████████▋ | 839/964 [00:29<00:04, 27.55 examples/s]
|
|
Tokenizing train dataset: 87%|████████▋ | 843/964 [00:29<00:04, 27.97 examples/s]
|
|
Tokenizing train dataset: 88%|████████▊ | 848/964 [00:29<00:03, 29.33 examples/s]
|
|
Tokenizing train dataset: 88%|████████▊ | 852/964 [00:29<00:03, 28.44 examples/s]
|
|
Tokenizing train dataset: 89%|████████▉ | 856/964 [00:29<00:03, 28.06 examples/s]
|
|
Tokenizing train dataset: 89%|████████▉ | 860/964 [00:29<00:03, 30.02 examples/s]
|
|
Tokenizing train dataset: 90%|████████▉ | 864/964 [00:29<00:03, 30.72 examples/s]
|
|
Tokenizing train dataset: 90%|█████████ | 868/964 [00:30<00:03, 29.51 examples/s]
|
|
Tokenizing train dataset: 90%|█████████ | 872/964 [00:30<00:03, 30.32 examples/s]
|
|
Tokenizing train dataset: 91%|█████████ | 877/964 [00:30<00:02, 30.99 examples/s]
|
|
Tokenizing train dataset: 91%|█████████▏| 881/964 [00:30<00:02, 28.78 examples/s]
|
|
Tokenizing train dataset: 92%|█████████▏| 885/964 [00:30<00:02, 29.77 examples/s]
|
|
Tokenizing train dataset: 92%|█████████▏| 889/964 [00:30<00:02, 29.93 examples/s]
|
|
Tokenizing train dataset: 93%|█████████▎| 894/964 [00:30<00:02, 33.09 examples/s]
|
|
Tokenizing train dataset: 93%|█████████▎| 898/964 [00:31<00:02, 32.29 examples/s]
|
|
Tokenizing train dataset: 94%|█████████▎| 902/964 [00:31<00:02, 29.18 examples/s]
|
|
Tokenizing train dataset: 94%|█████████▍| 906/964 [00:31<00:02, 27.76 examples/s]
|
|
Tokenizing train dataset: 95%|█████████▍| 911/964 [00:31<00:02, 25.78 examples/s]
|
|
Tokenizing train dataset: 95%|█████████▍| 915/964 [00:31<00:01, 27.39 examples/s]
|
|
Tokenizing train dataset: 95%|█████████▌| 918/964 [00:31<00:01, 25.83 examples/s]
|
|
Tokenizing train dataset: 96%|█████████▌| 923/964 [00:31<00:01, 27.93 examples/s]
|
|
Tokenizing train dataset: 96%|█████████▌| 926/964 [00:32<00:01, 27.69 examples/s]
|
|
Tokenizing train dataset: 97%|█████████▋| 931/964 [00:32<00:01, 31.04 examples/s]
|
|
Tokenizing train dataset: 97%|█████████▋| 935/964 [00:32<00:00, 31.84 examples/s]
|
|
Tokenizing train dataset: 98%|█████████▊| 941/964 [00:32<00:00, 29.31 examples/s]
|
|
Tokenizing train dataset: 98%|█████████▊| 946/964 [00:32<00:00, 30.67 examples/s]
|
|
Tokenizing train dataset: 99%|█████████▊| 951/964 [00:32<00:00, 27.05 examples/s]
|
|
Tokenizing train dataset: 99%|█████████▉| 956/964 [00:33<00:00, 28.73 examples/s]
|
|
Tokenizing train dataset: 99%|█████████▉| 959/964 [00:33<00:00, 27.11 examples/s]
|
|
Tokenizing train dataset: 100%|██████████| 964/964 [00:33<00:00, 29.50 examples/s]
|
|
Tokenizing train dataset: 100%|██████████| 964/964 [00:35<00:00, 27.32 examples/s]
|
|
|
|
Truncating train dataset: 0%| | 0/964 [00:00<?, ? examples/s]
|
|
Truncating train dataset: 100%|██████████| 964/964 [00:00<00:00, 4715.64 examples/s]
|
|
Truncating train dataset: 100%|██████████| 964/964 [00:00<00:00, 4545.39 examples/s]
|
|
2026-04-13 22:54:00,368 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,368 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,368 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,368 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,368 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,368 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,368 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,368 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,380 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,383 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,383 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,383 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,383 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,383 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,383 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,383 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,445 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,445 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,445 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,445 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,445 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,445 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,445 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,446 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,447 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,447 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,447 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,447 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,447 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,447 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,447 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,447 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,455 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,456 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,456 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,456 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,457 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,456 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,456 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,456 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,456 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,456 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,456 - INFO - Applying Liger kernels to model instance with model type: qwen3 with kwargs: {}
|
|
2026-04-13 22:54:00,458 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,457 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,459 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,459 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,460 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,461 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,468 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,469 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,471 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,471 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,471 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,471 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,471 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
2026-04-13 22:54:00,471 - INFO - Resuming trainer state from checkpoint: runs/dev/tw-8node-resume/checkpoint-64
|
|
wandb: WARNING The `run_name` is currently set to the same value as `TrainingArguments.output_dir`. If this was not intended, please specify a different run name by setting the `TrainingArguments.run_name` parameter.
|
|
wandb: Currently logged in as: ligeng-zhu to https://api.wandb.ai. Use `wandb login --relogin` to force relogin
|
|
wandb: creating run
|
|
wandb: Tracking run with wandb version 0.21.0
|
|
wandb: Run data is saved locally in runs/dev/tw-8node-resume/wandb/run-20260413_225408-tw-8node-resume
|
|
wandb: Run `wandb offline` to turn off syncing.
|
|
wandb: Resuming run runs/dev/tw-8node-resume
|
|
wandb: ⭐️ View project at https://wandb.ai/ligeng-zhu/ThreadWeaver
|
|
wandb: 🚀 View run at https://wandb.ai/ligeng-zhu/ThreadWeaver/runs/tw-8node-resume
|
|
A ConfigError was raised whilst setting the number of model parameters in Weights & Biases config.
|
|
|
|
0%| | 0/64 [00:00<?, ?it/s]
|
|
|
|
|
|
0%| | 0/64 [00:00<?, ?it/s]
|
|
0%| | 0/64 [00:00<?, ?it/s]
|