© 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. Residuals
    9. 2-RES-SM
    11. 26170
    Prev
    Next
    INDEX
    Explanations

    legal references and case law citations

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-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_resid_post
    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
    419
    +0.18
    1.0%
    237
    +0.14
    0.8%
    294
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    506
    +0.18
    0.11
    237
    +0.14
    0.14
    294
    +0.14
    0.07
    Negative Logits
     historian
    -1.35
    :`
    -1.34
     Archae
    -1.33
    opause
    -1.32
     Census
    -1.32
     Wikipedia
    -1.30
    cex
    -1.29
     ERA
    -1.27
    ada
    -1.26
     incomp
    -1.24
    POSITIVE LOGITS
     hostage
    1.55
    ĥ½
    1.48
    Willi
    1.47
    imide
    1.46
    ¬
    1.41
    ever
    1.41
     punches
    1.41
    _->
    1.38
    ´
    1.37
    ¿½
    1.36
    Activations Density 7.295%

    No Known Activations