🦙📚 LlamaTales - a ivnle Collection

ivnle 's Collections

updated Nov 11

Datasets and checkpoints from the paper LlamaTales: Studying the Effects of Developmentally Inspired Training Data on Small Language Models

Upvote

ivnle/llamatales-gre-70b

Viewer • Updated Nov 8 • 2M • 47

Note Short stories generated by `nvidia/Llama-3.1-Nemotron-70B-Instruct`.
ivnle/llamatales-jr-70b

Viewer • Updated Nov 9 • 3.56M • 39

Note Children's stories generated by `nvidia/Llama-3.1-Nemotron-70B-Instruct`.
ivnle/llamatales-gre

Viewer • Updated Oct 21 • 2.02M • 99

Note Short stories generated by `meta-llama/Llama-3.1-8B-Instruct`.
ivnle/llamatales-jr

Viewer • Updated Oct 22 • 3.59M • 79

Note Children's stories generated by `meta-llama/Llama-3.1-8B-Instruct`.
ivnle/tinystories

Viewer • Updated Oct 22 • 4.97M • 40

Note Source: https://huggingface.co/datasets/roneneldan/TinyStories/blob/main/TinyStories_all_data.tar.gz
ivnle/fineweb

Viewer • Updated Oct 22 • 2.03M • 42

Note 1B token sample of FineWeb-Edu https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu Model checkpoints below. Naming format is [training data]-[layers]-[hidden size]-[heads]-[non-embedding parameter count].
ivnle/llamatales_gre_8b-lay8-hs512-hd8-33M

Text Generation • Updated Nov 14 • 12
ivnle/llamatales_gre_8b-lay8-hs384-hd6-18M

Text Generation • Updated Nov 14 • 14
ivnle/llamatales_gre_8b-lay4-hs384-hd6-9M

Text Generation • Updated Nov 14 • 10
ivnle/llamatales_gre_8b-lay4-hs128-hd2-1M

Text Generation • Updated Nov 14 • 12
ivnle/llamatales_gre_8b-lay2-hs128-hd2-524K

Text Generation • Updated Nov 14 • 11
ivnle/llamatales_gre_8b-lay1-hs128-hd2-262K

Text Generation • Updated Nov 14 • 8
ivnle/fineweb-lay8-hs512-hd8-33M

Text Generation • Updated Nov 14 • 10
ivnle/fineweb-lay8-hs384-hd6-18M

Text Generation • Updated Nov 14 • 8
ivnle/fineweb-lay4-hs384-hd6-9M

Text Generation • Updated Nov 14 • 7
ivnle/fineweb-lay4-hs128-hd2-1M

Text Generation • Updated Nov 14 • 6
ivnle/fineweb-lay2-hs128-hd2-524K

Text Generation • Updated Nov 14 • 10
ivnle/fineweb-lay1-hs128-hd2-262K

Text Generation • Updated Nov 14 • 12
ivnle/llamatales_jr_8b-lay8-hs512-hd8-33M

Text Generation • Updated Nov 14 • 264
ivnle/llamatales_jr_8b-lay8-hs384-hd6-18M

Text Generation • Updated Nov 14 • 12
ivnle/llamatales_jr_8b-lay4-hs384-hd6-9M

Text Generation • Updated Nov 14 • 10
ivnle/llamatales_jr_8b-lay4-hs128-hd2-1M

Text Generation • Updated Nov 14 • 13
ivnle/llamatales_jr_8b-lay2-hs128-hd2-524K

Text Generation • Updated Nov 14 • 8
ivnle/llamatales_jr_8b-lay1-hs128-hd2-262K

Text Generation • Updated Nov 14 • 7
ivnle/tinystories-lay8-hs512-hd8-33M

Text Generation • Updated Nov 14 • 9
ivnle/tinystories-lay8-hs384-hd6-18M

Text Generation • Updated Nov 14 • 12
ivnle/tinystories-lay4-hs384-hd6-9M

Text Generation • Updated Nov 14 • 10
ivnle/tinystories-lay4-hs128-hd2-1M

Text Generation • Updated Nov 14 • 8
ivnle/tinystories-lay2-hs128-hd2-524K

Text Generation • Updated Nov 14 • 7
ivnle/tinystories-lay1-hs128-hd2-262K

Text Generation • Updated Nov 14 • 11

Upvote