© 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. 4-RES-SM
    11. 8975
    Prev
    Next
    INDEX
    Explanations

    repetitive phrases emphasizing precision or certainty

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    +0.12
    0.7%
    458
    +0.12
    0.6%
    105
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    458
    +0.12
    0.02
    105
    +0.12
    0.05
    75
    +0.11
    0.01
    Negative Logits
    ĥ½
    -3.09
    ij
    -2.98
    ¾
    -2.71
    »
    -2.64
    Ī
    -2.56
    ĵ
    -2.55
    ľĵ
    -2.52
    ĨĴ
    -2.49
    į
    -2.49
    Ģ
    -2.47
    POSITIVE LOGITS
     what
    2.04
     analogous
    2.03
     why
    1.74
     identical
    1.63
     indistinguishable
    1.63
     opposite
    1.59
     same
    1.51
     unchanged
    1.49
     equal
    1.48
     how
    1.46
    Activations Density 0.340%

    No Known Activations