© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 22475
    Prev
    Next
    INDEX
    Explanations

    references to roads and related infrastructure

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_mlp_out
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    278
    +0.14
    0.8%
    327
    +0.11
    0.6%
    156
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    278
    +0.14
    0.01
    327
    +0.11
    0.04
    23
    +0.11
    0.02
    Negative Logits
    ĵ
    -3.27
    Ł
    -3.16
    º
    -3.14
    ĸ
    -3.06
    ²
    -2.94
    ¹
    -2.89
    į
    -2.84
    §
    -2.84
    ĻĤ
    -2.84
    ¿½
    -2.83
    POSITIVE LOGITS
     truth
    1.39
     richest
    1.39
     \(
    1.35
    roscopic
    1.34
     representative
    1.33
     unbiased
    1.32
     bin
    1.31
     receiver
    1.30
     abundant
    1.29
     inventor
    1.29
    Activations Density 0.587%

    No Known Activations