© 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. 24971
    Prev
    Next
    INDEX
    Explanations

    expressions of gratitude

    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
    376
    +0.14
    0.8%
    478
    +0.14
    0.8%
    296
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    147
    +0.14
    0.02
    317
    +0.14
    0.02
    42
    +0.11
    0.01
    Negative Logits
    ĭ
    -3.07
    Ģ
    -2.97
    Ŀ
    -2.95
    ħ
    -2.94
    £
    -2.82
    Ļª
    -2.75
    »¿
    -2.73
    ij
    -2.72
    ĩ
    -2.70
    °
    -2.68
    POSITIVE LOGITS
     neon
    1.75
    xygen
    1.58
     moderator
    1.52
     omissions
    1.40
     scholar
    1.39
    printStackTrace
    1.38
     careful
    1.38
     urgency
    1.36
     retrospect
    1.36
     crim
    1.36
    Activations Density 0.145%

    No Known Activations