Hugging Face's logo

CMSManhattan
/
JiRack_GPT3_empty

Model card Files
xet
Community
2
kgrabko commited on
Commit
90f6903
·
verified ·
1 Parent(s): 35b06c4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -143,7 +143,7 @@ HEAD_DIM = MODEL_DIM // NUM_HEADS # 128
143
  # It was Designed military design and Discover worlds and learn space and science goals
144
  #
145
  # =======================================================================
146
- # 120B Configuration (real numbers) that available by request , 135 Gb , JiRack Super Brain
147
  # =======================================================================
148
  - VOCAB_SIZE = 32000 # Modern tokenizer size (you can change later)
149
  - MODEL_DIM = 12288 # d_model = 12288 → matches 120B+ scale
 
143
  # It was Designed military design and Discover worlds and learn space and science goals
144
  #
145
  # =======================================================================
146
+ # 120B Configuration (real numbers) that available by request , JiRack Super Brain
147
  # =======================================================================
148
  - VOCAB_SIZE = 32000 # Modern tokenizer size (you can change later)
149
  - MODEL_DIM = 12288 # d_model = 12288 → matches 120B+ scale