Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

Compactbot
/
ldt-10m

Text Generation
Safetensors
English
ldt
tiny
tiny-lm
tiny-model
slm
SLM
small-language-model
from-scratch
llama
Model card Files Files and versions
xet
Community
15
ldt-10m
42 MB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 20 commits
Compactbot's picture
Compactbot
Update card to v8 (current weights): 79,375 steps / 2.60B tokens (hits target), val 3.63. Honest eval: 40-sample sweep shows mean loop frac 0.219 (up from v7's 0.167) and 14/40 degenerate โ€” more tokens did not buy coherence. Replaces the stale v7-only card that described the weights as 'well short of 2.6B'.
1cc0879 verified about 18 hours ago
  • .gitattributes
    1.52 kB
    initial commit 5 days ago
  • README.md
    5.58 kB
    Update card to v8 (current weights): 79,375 steps / 2.60B tokens (hits target), val 3.63. Honest eval: 40-sample sweep shows mean loop frac 0.219 (up from v7's 0.167) and 14/40 degenerate โ€” more tokens did not buy coherence. Replaces the stale v7-only card that described the weights as 'well short of 2.6B'. about 18 hours ago
  • config.json
    306 Bytes
    Update config to v7 checkpoint (step 30000, best_val 3.6892) 4 days ago
  • model.py
    5.62 kB
    Self-contained LDT model definition (#2) 5 days ago
  • model.safetensors
    41.1 MB
    xet
    v8: continued training to 2.6B tokens (from 1.51B). Improved coherence on narrative prompts. Same architecture (10,052,864 params, 5L d320 5-head SwiGLU RoPE). (#14) about 19 hours ago
  • tokenizer.json
    830 kB
    Byte-level BPE tokenizer (vocab 12288) (#3) 5 days ago
  • tokenizer_config.json
    250 Bytes
    Tokenizer config (#5) 5 days ago