metadata
license: mit
pipeline_tag: text-generation
library_name: transformers
language:
- en
- am
- ar
- as
- az
- be
- bg
- bn
- br
- bs
- ca
- cs
- cy
- da
- de
- el
- eo
- es
- et
- eu
- fa
- ff
- fi
- fr
- fy
- ga
- gd
- gl
- gn
- gu
- ha
- he
- hi
- hr
- ht
- hu
- hy
- id
- ig
- is
- it
- ja
- jv
- ka
- kk
- km
- kn
- ko
- ku
- ky
- la
- lg
- li
- ln
- lo
- lt
- lv
- mg
- mk
- ml
- mn
- mr
- ms
- my
- ne
- nl
- 'no'
- ns
- om
- or
- pa
- pl
- ps
- pt
- qu
- rm
- ro
- ru
- sa
- si
- sc
- sd
- sk
- sl
- so
- sq
- sr
- ss
- su
- sv
- sw
- ta
- te
- th
- tl
- tn
- tr
- ug
- uk
- ur
- uz
- vi
- wo
- xh
- yi
- yo
- zu
datasets:
- ontocord/fineweb-permissive-multilingual-2m
- distily/c4_multilingual_1M
- data-silence/sumnews
- xu-song/cc100-samples
- badrex/llm-emoji-dataset
- fblgit/simple-math
- Gusarich/math-expressions-1m
- neuralwork/arxiver
- christopher/rosetta-code
- nampdn-ai/tiny-codes
- JeanKaddour/minipile
- NousResearch/hermes-function-calling-v1
- simplescaling/s1K-1.1
- mlabonne/open-perfectblend
- allenai/tulu-3-sft-mixture
- rombodawg/Everything_Instruct_Multilingual
- open-r1/OpenR1-Math-220k
- open-thoughts/OpenThoughts-114k
- cognitivecomputations/dolphin-r1
- simplescaling/s1K-1.1
tags:
- chat
- core
- base
- instruct
- reason
tangled-alpha-0.10-core
time python -B prepare_core_datasets.py
i=0, min_len=0, max_len=1073741824, block_size=1025, chunk_size=16400000, len(dataset)=5146620, len(dataset) * block_size=5275285500
Total number of tokens in the optimized dataset '../core-data-0-0-1073741824-1025-16000' is 5275285500
i=1, min_len=1025, max_len=2049, block_size=2049, chunk_size=16392000, len(dataset)=309838, len(dataset) * block_size=634858062
Total number of tokens in the optimized dataset '../core-data-1-1025-2049-2049-8000' is 634858062
i=2, min_len=2049, max_len=4097, block_size=4097, chunk_size=16388000, len(dataset)=113843, len(dataset) * block_size=466414771
Total number of tokens in the optimized dataset '../core-data-2-2049-4097-4097-4000' is 466414771
i=3, min_len=4097, max_len=8193, block_size=8193, chunk_size=16386000, len(dataset)=56713, len(dataset) * block_size=464649609
Total number of tokens in the optimized dataset '../core-data-3-4097-8193-8193-2000' is 464649609
i=4, min_len=8193, max_len=16385, block_size=16385, chunk_size=16385000, len(dataset)=37406, len(dataset) * block_size=612897310
Total number of tokens in the optimized dataset '../core-data-4-8193-16385-16385-1000' is 612897310
i=5, min_len=16385, max_len=32769, block_size=32769, chunk_size=16384500, len(dataset)=12737, len(dataset) * block_size=417378753
Total number of tokens in the optimized dataset '../core-data-5-16385-32769-32769-500' is 417378753
i=6, min_len=32769, max_len=65537, block_size=65537, chunk_size=16384250, len(dataset)=2824, len(dataset) * block_size=185076488
Total number of tokens in the optimized dataset '../core-data-6-32769-65537-65537-250' is 185076488
i=7, min_len=65537, max_len=131073, block_size=131073, chunk_size=16384125, len(dataset)=634, len(dataset) * block_size=83100282
Total number of tokens in the optimized dataset '../core-data-7-65537-131073-131073-125' is 83100282
real 292m54.341s
user 2118m1.154s
sys 12m2.746s
20G tangled-alpha-0.9-core/core-data-0-0-1073741824-1025-16000
2.4G tangled-alpha-0.9-core/core-data-1-1025-2049-2049-8000
1.8G tangled-alpha-0.9-core/core-data-2-2049-4097-4097-4000
1.8G tangled-alpha-0.9-core/core-data-3-4097-8193-8193-2000
2.3G tangled-alpha-0.9-core/core-data-4-8193-16385-16385-1000
1.6G tangled-alpha-0.9-core/core-data-5-16385-32769-32769-500
709M tangled-alpha-0.9-core/core-data-6-32769-65537-65537-250
321M tangled-alpha-0.9-core/core-data-7-65537-131073-131073-125
CUDA_VISIBLE_DEVICES=0 CUDA_LAUNCH_BLOCKING=0 PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True litgpt pretrain --config pretrain_core_model_0.yaml
Seed set to 23
Time to instantiate model: 0.21 seconds.
Total parameters: 302,023,168
Verifying settings ...
Measured TFLOPs: 55520.94
Epoch 1 | iter 64 step 1 | loss train: 11.982, val: n/a | iter time: 409.55 ms (step) remaining time: 4 days, 17:45:21
Epoch 1 | iter 128 step 2 | loss train: 11.980, val: n/a | iter time: 354.46 ms (step) remaining time: 3 days, 15:01:16
Epoch 1 | iter 192 step 3 | loss train: 11.980, val: n/a | iter time: 353.67 ms (step) remaining time: 3 days, 5:46:03
Epoch 1 | iter 256 step 4 | loss train: 11.980, val: n/a | iter time: 354.11 ms (step) remaining time: 3 days, 1:05:26
Epoch 1 | iter 320 step 5 | loss train: 11.978, val: n/a | iter time: 358.28 ms (step) remaining time: 2 days, 22:21:45
Epoch 1 | iter 384 step 6 | loss train: 11.974, val: n/a | iter time: 356.21 ms (step) remaining time: 2 days, 20:33:55
Epoch 1 | iter 448 step 7 | loss train: 11.964, val: n/a | iter time: 357.42 ms (step) remaining time: 2 days, 19:15:59
Epoch 1 | iter 512 step 8 | loss train: 11.956, val: n/a | iter time: 355.74 ms (step) remaining time: 2 days, 18:16:43
Epoch 1 | iter 576 step 9 | loss train: 11.937, val: n/a | iter time: 356.05 ms (step) remaining time: 2 days, 17:28:34
Epoch 1 | iter 640 step 10 | loss train: 11.929, val: n/a | iter time: 356.68 ms (step) remaining time: 2 days, 16:49:58
# ...
Backup wandb:
mv wandb wandb-pretrain-core-0
Copy config:
cp ../config-0.json ../out/pretrain-core-0/final/config.json
Chat with model:
CUDA_VISIBLE_DEVICES=0 CUDA_LAUNCH_BLOCKING=0 PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True litgpt chat ../out/pretrain-core-0/final
CUDA_VISIBLE_DEVICES=0 CUDA_LAUNCH_BLOCKING=0 PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True time litgpt evaluate --tasks 'leaderboard' --out_dir '../evaluate/pretrain-core-0/leaderboard/' --batch_size '4' --dtype 'bfloat16' '../out/pretrain-core-0/final'