© 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. 10502
    Prev
    Next
    INDEX
    Explanations

    phrases related to evaluating and assessing a situation or performance

    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
    348
    +0.20
    1.1%
    25
    +0.17
    1.0%
    466
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    140
    +0.20
    0.09
    418
    +0.17
    0.04
    289
    +0.16
    0.04
    Negative Logits
     trillion
    -1.96
    []{
    -1.79
     exports
    -1.75
     empire
    -1.73
    !’
    -1.70
     billion
    -1.69
    !”
    -1.62
     flavor
    -1.56
     Million
    -1.51
     extinct
    -1.50
    POSITIVE LOGITS
    Ļª
    3.15
     clinician
    2.94
     therapists
    2.91
     patients
    2.86
     clinicians
    2.85
     patient
    2.84
     therapist
    2.74
     physicians
    2.69
    ĥ½
    2.67
    ľĵ
    2.67
    Activations Density 2.993%

    No Known Activations