© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 28227
    Prev
    Next
    INDEX
    Explanations

    terms related to government structure and policy impact

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    6
    +0.13
    0.8%
    421
    +0.13
    0.7%
    156
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    347
    +0.13
    0.08
    365
    +0.13
    0.08
    455
    +0.12
    0.05
    Negative Logits
    erge
    -1.58
    eting
    -1.45
    izabeth
    -1.41
     longer
    -1.41
    anges
    -1.39
    ---|---
    -1.35
    .
    -1.33
     unde
    -1.31
    chle
    -1.31
     %.
    -1.29
    POSITIVE LOGITS
    ously
    1.81
     however
    1.69
     investigators
    1.55
     raised
    1.46
    ĻĤ
    1.35
     reactions
    1.32
    iously
    1.31
     indications
    1.31
    breviations
    1.29
     representatives
    1.28
    Activations Density 3.085%

    No Known Activations