InternLM/InternBootcamp
1
0
Fork 0
mirror of https://github.com/InternLM/InternBootcamp.git synced 2026-04-22 16:49:04 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

refactor(data_pipeline): optimize data generation pipeline; add multiple preset configurations for data generation

Browse source
This commit is contained in:
chenyongkang 2025-08-26 18:40:21 +08:00
parent 1a8477c8d8
commit 8d493b35a0
2160 changed files with 69199 additions and 154 deletions
Download patch file Download diff file Expand all files Collapse all files

78
examples/pipelines/all_configs/Cantonandfairytale_test.json Normal file
View file

@ -0,0 +1,78 @@
[
{
"max_n": 5000000000,
"max_m": 250000000
},
{
"max_n": 5000000000000000,
"max_m": 250000000000000
},
{
"max_n": 10000000,
"max_m": 5000000
},
{
"max_n": 500000000000000000,
"max_m": 25000000000000000
},
{
"max_n": 1000000000,
"max_m": 500000000
},
{
"max_n": 50000000000,
"max_m": 2500000000
},
{
"max_n": 1000000000000000,
"max_m": 500000000000000
},
{
"max_n": 500000000,
"max_m": 25000000
},
{
"max_n": 10000000000,
"max_m": 5000000000
},
{
"max_n": 100000000000000,
"max_m": 50000000000000
},
{
"max_n": 100,
"max_m": 100
},
{
"max_n": 500000,
"max_m": 1000
},
{
"max_n": 5000000,
"max_m": 100000
},
{
"max_n": 50000000,
"max_m": 2500000
},
{
"max_n": 10000000000000,
"max_m": 5000000000000
},
{
"max_n": 5000000000000,
"max_m": 250000000000
},
{
"max_n": 500000000000000,
"max_m": 25000000000000
},
{
"max_n": 10000000000000000,
"max_m": 5000000000000000
},
{
"max_n": 100000000,
"max_m": 50000000
}
]