Datasets:

Chamaka8
/

serendip-cpt-sinhala

Tasks:

Text Generation

Modalities:

Formats:

Languages:

Size:

Tags:

continual-pretraining

Libraries:

License:

Dataset card Data Studio Files Files and versions

serendip-cpt-sinhala

21.6 GB

1 contributor

History: 21 commits

Chamaka8's picture

Add full dataset card with statistics, schema, usage examples

d9bd91a verified 2 months ago

data
Add clean CPT shards, rebuild scripts, and evaluation proof files 5 months ago
runpod
Add training scripts for CPT, SFT-LoRA, DPO and RunPod support 5 months ago
train_scripts
Add clean CPT shards, rebuild scripts, and evaluation proof files 5 months ago
.gitattributes

2.58 kB
Add clean CPT shards, rebuild scripts, and evaluation proof files 5 months ago
.gitignore

63 Bytes
Add clean CPT shards, rebuild scripts, and evaluation proof files 5 months ago
README.md

9.64 kB
Add full dataset card with statistics, schema, usage examples 2 months ago
compare_small_vs_cpt.py

3.65 kB
Add clean CPT shards, rebuild scripts, and evaluation proof files 5 months ago
mini_run.py

6.9 kB
Add training scripts for CPT, SFT-LoRA, DPO and RunPod support 5 months ago
test_base.py

850 Bytes
Add clean CPT shards, rebuild scripts, and evaluation proof files 5 months ago
test_cpt.py

856 Bytes
Add clean CPT shards, rebuild scripts, and evaluation proof files 5 months ago