k/llm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • 98282675b0 Fixed embeding mistake main k 2026-03-05 15:13:34 -05:00
  • 8d8fb8c212 experimental "lin" blocks insted of attention sparely k 2026-02-27 09:13:24 -05:00
  • 89c9d01cb8 More training with less heads k 2026-02-27 09:13:01 -05:00
  • dc231ae703 fixed bos token being prepended twice k 2026-02-27 09:10:34 -05:00
  • a0cd98876c added gitignore k 2026-01-13 21:11:32 -05:00
  • 0537a5df64 changed chat dataset. k 2026-01-09 17:30:34 -05:00
  • c78a31362a set to gpt2 hyprs k 2026-01-09 12:45:01 -05:00
  • 496916f428 added fine-tuning k 2026-01-07 13:01:06 -05:00
  • 121640bab6 updated hypr for my gpu k 2026-01-07 12:59:44 -05:00
  • 6f037c4a9a Quick training script k 2026-01-07 02:14:09 -05:00
  • 7f25dff1d1 Fix errors k 2026-01-07 02:13:08 -05:00
  • 007c96e91b Simple log functions k 2026-01-07 01:25:47 -05:00
  • 6daa8ec46c Added code to generate training batches k 2026-01-07 01:15:18 -05:00
  • 229c564811 CosineAnnealing with optimizer Group k 2026-01-07 00:26:04 -05:00
  • 478010c8cc added Positional encodeings k 2026-01-06 21:38:12 -05:00
  • 3b590b3ce7 added dropout to ffn k 2026-01-06 21:26:51 -05:00
  • 957aad2239 Implimented Transformer(decode only) k 2026-01-06 21:26:24 -05:00
  • 23f62c7e64 Implimented TransformerBlock k 2026-01-06 19:53:37 -05:00
  • 77aa0de0eb Implimented MultiHeadAttention k 2026-01-06 19:41:12 -05:00
  • c4e5e332ba fixed cast in ffn k 2026-01-06 19:40:45 -05:00
  • d6b9f45fcc Implimented Feed Forward Netwok k 2026-01-06 18:31:04 -05:00