© 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. 25309
    Prev
    Next
    INDEX
    Explanations

    references to personal pronouns, particularly "me"

    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
    189
    +0.16
    0.9%
    365
    +0.12
    0.7%
    444
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    365
    +0.16
    0.06
    363
    +0.12
    0.06
    338
    +0.11
    0.07
    Negative Logits
    uclear
    -1.79
    ellation
    -1.60
    ?’
    -1.52
     ‘
    -1.51
    ¿½
    -1.50
     edge
    -1.42
     ’
    -1.42
    edge
    -1.41
     equivalent
    -1.40
     QR
    -1.37
    POSITIVE LOGITS
    xico
    2.63
    "}](#
    1.64
    ÅĽci
    1.45
    chle
    1.44
    zzo
    1.42
    cause
    1.42
    hof
    1.42
    burg
    1.40
    dorff
    1.38
    rass
    1.37
    Activations Density 0.754%

    No Known Activations