© 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. 27145
    Prev
    Next
    INDEX
    Explanations

    the word "feel" and its variations, indicating emotional responses

    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
    144
    +0.14
    0.8%
    32
    +0.13
    0.7%
    203
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    32
    +0.14
    0.02
    144
    +0.13
    0.02
    203
    +0.13
    0.02
    Negative Logits
    poral
    -1.53
    sis
    -1.46
    than
    -1.41
    thood
    -1.38
    folio
    -1.37
    tel
    -1.35
    ESULT
    -1.35
    apses
    -1.35
    tum
    -1.34
    death
    -1.32
    POSITIVE LOGITS
    ynman
    2.21
    hler
    2.19
    ather
    2.13
    inberg
    1.94
    instein
    1.94
    atured
    1.90
    aturing
    1.89
    edly
    1.87
    isty
    1.87
    eling
    1.83
    Activations Density 0.175%

    No Known Activations