© 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. 29195
    Prev
    Next
    INDEX
    Explanations

    the presence and frequency of specific numerical values and operations or commands

    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
    70
    +0.12
    0.7%
    256
    +0.11
    0.6%
    150
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    273
    +0.12
    0.02
    150
    +0.11
    0.03
    18
    +0.11
    0.03
    Negative Logits
    ĻĤ
    -1.92
    ij
    -1.82
    Īĺ
    -1.74
    Ļª
    -1.71
    Ĺ
    -1.67
    £
    -1.66
    Į
    -1.65
    ĨĴ
    -1.63
    ”?
    -1.59
    optera
    -1.59
    POSITIVE LOGITS
    vert
    1.69
    face
    1.52
     presentation
    1.46
     apologies
    1.39
    ringer
    1.38
     degrees
    1.36
     discussion
    1.35
     COPYRIGHT
    1.34
    pone
    1.32
    writing
    1.32
    Activations Density 0.218%

    No Known Activations