© 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. 2-RES-SM
    11. 28035
    Prev
    Next
    INDEX
    Explanations

    mentions of dates and accompanying numerical information

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    258
    +0.20
    1.1%
    23
    +0.15
    0.9%
    369
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    258
    +0.20
    0.17
    23
    +0.15
    0.13
    321
    +0.13
    0.09
    Negative Logits
    azzo
    -1.72
     quarters
    -1.69
    tees
    -1.69
    ister
    -1.57
    mittee
    -1.55
    ivel
    -1.53
     course
    -1.53
    guy
    -1.52
    ème
    -1.49
    ius
    -1.48
    POSITIVE LOGITS
    Ĥ
    2.94
    Ī
    2.73
    ĥ½
    2.72
    ½
    2.61
    ©
    2.57
    ¯
    2.51
    ĸ
    2.49
    į
    2.47
    º
    2.46
    §
    2.42
    Activations Density 1.084%

    No Known Activations