© 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. 32190
    Prev
    Next
    INDEX
    Explanations

    the presence of a specialized command or function related to "access."

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    343
    +0.15
    0.9%
    410
    +0.12
    0.7%
    205
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    343
    +0.15
    0.03
    67
    +0.12
    0.02
    86
    +0.10
    0.02
    Negative Logits
    iture
    -2.01
    ESULT
    -1.94
     other
    -1.38
     recent
    -1.37
    aintiff
    -1.35
    eing
    -1.35
    itures
    -1.34
    -1.32
     exact
    -1.28
    bst
    -1.27
    POSITIVE LOGITS
    hern
    1.92
    iously
    1.86
    ellular
    1.84
    ubes
    1.79
    ruz
    1.79
    hes
    1.69
    ramento
    1.67
    hens
    1.60
    oustic
    1.58
    mas
    1.55
    Activations Density 0.098%

    No Known Activations