© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 24093
    Prev
    Next
    INDEX
    Explanations

    mentions of geographical locations, specifically coastal areas

    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
    321
    +0.14
    0.8%
    43
    +0.14
    0.8%
    119
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    321
    +0.14
    0.02
    43
    +0.14
    0.02
    220
    +0.13
    0.01
    Negative Logits
    nia
    -1.91
    Ķ
    -1.69
    Į
    -1.67
    Ĥ
    -1.62
    >(
    -1.59
     tears
    -1.56
    chie
    -1.55
    ¶
    -1.53
    odal
    -1.52
    >&
    -1.50
    POSITIVE LOGITS
    line
    2.02
    liner
    1.93
    break
    1.84
    boats
    1.79
    stones
    1.74
    boat
    1.74
    roads
    1.69
    piece
    1.66
    pieces
    1.65
    lights
    1.63
    Activations Density 0.084%

    No Known Activations