© 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. 2-RES-SM
    11. 13458
    Prev
    Next
    INDEX
    Explanations

    phrases and sentences indicating negation or absence

    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
    320
    +0.12
    0.6%
    267
    +0.11
    0.6%
    39
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    206
    +0.12
    0.05
    267
    +0.11
    0.06
    39
    +0.11
    0.05
    Negative Logits
    Ļª
    -2.33
    ±
    -2.10
    ĨĴ
    -1.82
    ľ
    -1.73
    ungen
    -1.65
    »¿
    -1.60
    ļ
    -1.58
     perspective
    -1.56
    ĸ´
    -1.54
    ĻĤ
    -1.52
    POSITIVE LOGITS
    oriously
    2.10
    necessarily
    2.05
     anymore
    1.95
     necessarily
    1.82
    yet
    1.80
    iq
    1.69
    sov
    1.65
    inherit
    1.57
    orious
    1.57
    exist
    1.52
    Activations Density 0.316%

    No Known Activations