© 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 EvalsExportsAPI 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. 26313
    Prev
    Next
    INDEX
    Explanations

    statistical methods and analyses concerning medical research data

    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
    467
    +0.12
    0.6%
    134
    +0.11
    0.6%
    42
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.12
    0.04
    8
    +0.11
    0.02
    498
    +0.11
    0.04
    Negative Logits
     nature
    -1.67
     soul
    -1.67
    roots
    -1.65
     queer
    -1.57
     fucking
    -1.52
     devil
    -1.51
     finest
    -1.50
     fortunes
    -1.49
     cher
    -1.48
     bounty
    -1.48
    POSITIVE LOGITS
    EMA
    1.83
    )^[@
    1.74
    EC
    1.71
    etic
    1.70
    edin
    1.68
    STEM
    1.66
     implanted
    1.59
    fecture
    1.57
    rectomy
    1.55
    ANC
    1.53
    Activations Density 0.459%

    No Known Activations