© 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. 4-RES-SM
    11. 8864
    Prev
    Next
    INDEX
    Explanations

    measurements and comparisons of chemical concentrations and related values in scientific contexts

    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
    498
    +0.14
    0.8%
    240
    +0.13
    0.7%
    268
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    262
    +0.14
    0.03
    498
    +0.13
    0.06
    269
    +0.11
    0.05
    Negative Logits
    ¥
    -4.68
    ½
    -4.36
    ¼
    -4.35
    ŀ
    -4.32
    ¦
    -4.30
    ľĵ
    -4.25
    ¡
    -4.24
    £
    -4.18
    ļ
    -4.17
    ª
    -4.08
    POSITIVE LOGITS
    creen
    2.94
    heet
    2.82
    hift
    2.82
    hip
    2.54
    hips
    2.47
    ource
    2.47
    plit
    2.41
    heets
    2.38
    ocial
    2.32
    kip
    2.30
    Activations Density 0.312%

    No Known Activations