© 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. 27199
    Prev
    Next
    INDEX
    Explanations

    specific references to institutions and formal organizations

    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
    156
    +0.16
    0.9%
    302
    +0.12
    0.7%
    450
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    57
    +0.16
    0.08
    373
    +0.12
    0.06
    450
    +0.11
    0.06
    Negative Logits
    §
    -1.88
    ¿½
    -1.87
    ĸ
    -1.83
    gets
    -1.68
    º
    -1.67
    ¬
    -1.65
    ½
    -1.64
    ÃĥÃĤÃĥÃĤ
    -1.60
    ģ
    -1.60
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    -1.58
    POSITIVE LOGITS
     Agent
    1.52
     Member
    1.46
     portrait
    1.44
     Ship
    1.38
    eks
    1.34
     dock
    1.33
     Bund
    1.30
     downs
    1.30
     default
    1.29
     autom
    1.28
    Activations Density 3.167%

    No Known Activations