๐ In a Training Loop
foss
foss22
AI & ML interests
None yet
Recent Activity
liked a model 9 days ago
erectbranch/sam3-gguf updated a model 18 days ago
foss22/NanoZaya-3B-A35M liked a model 18 days ago
ngquocvinh/AliceAI-T5-35B-A0.6B-GGUFOrganizations
Reproducibility appreciated
1
#1 opened 28 days ago
by
foss22
Extend vocab with songs phrases: 32k => 65k tokens
#7 opened 29 days ago
by
foss22
1025 - ~2000 token text vs 2-3 words
๐ 1
3
#2 opened about 1 month ago
by
foss22
Agentic 0,07 B LM tasks + complete list of use-cases for Vintage LMs
3
#3 opened about 1 month ago
by
foss22
77 million parameters is equal to 0.077 billion, not 0.007
โค๏ธ 1
#2 opened about 1 month ago
by
foss22
Per-profession vocab
๐ 1
6
#4 opened about 2 months ago
by
foss22
RedBook tokenizier. Biodiversity-consetvation inspired
#6 opened about 1 month ago
by
foss22
Tokens indicate unbalanced subsets, that do not represent general English
#8 opened about 2 months ago
by
foss22
ร |ฤฏ|ร|ยง|ยฉ
1
#7 opened about 2 months ago
by
foss22
Remove ---- .... === ### dupes from datasets and tokenizer
๐ 1
4
#3 opened about 2 months ago
by
foss22
Capitalization not normalized
#6 opened about 2 months ago
by
foss22
When capital letter matters?
#5 opened about 2 months ago
by
foss22
Page numbers dominate 645 num tokens
#4 opened about 2 months ago
by
foss22
Collect test inference or no value of collecting?
6
#1 opened about 2 months ago
by
foss22
29 of 798 Wikipedia 1900-1990s neologisms
5
#1 opened about 2 months ago
by
foss22
Questions in 454k vs 783k without questions
1
#5 opened about 2 months ago
by
foss22
Redundancy
1
#4 opened about 2 months ago
by
foss22
Under-filtering.
1
#3 opened about 2 months ago
by
foss22
Over-filtering problem with current banned.txt
1
#2 opened about 2 months ago
by
foss22
F16 NaN in 10 steps workaround
1
#2 opened about 2 months ago
by
foss22