© 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. 24043
    Prev
    Next
    INDEX
    Explanations

    references to specific geographic locations or entities in text

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.16
    0.9%
    410
    +0.13
    0.7%
    80
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.16
    0.03
    80
    +0.13
    0.02
    496
    +0.12
    0.02
    Negative Logits
     sexes
    -1.66
    ?'
    -1.66
    ,'"
    -1.62
     ]$
    -1.58
     overweight
    -1.55
     Ethiop
    -1.53
    ,'
    -1.50
    ]^{\
    -1.48
     genetically
    -1.47
     \;
    -1.46
    POSITIVE LOGITS
    shire
    2.31
    ière
    1.86
    ski
    1.82
    imore
    1.81
    vier
    1.81
    thouse
    1.73
    illac
    1.66
    wagen
    1.66
    kowski
    1.64
    istics
    1.62
    Activations Density 0.108%

    No Known Activations