© 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. 24171
    Prev
    Next
    INDEX
    Explanations

    the plural form of the word "he."

    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
    255
    +0.11
    0.7%
    180
    +0.11
    0.7%
    101
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    58
    +0.11
    0.02
    378
    +0.11
    0.02
    380
    +0.11
    0.02
    Negative Logits
     preponderance
    -1.94
     weak
    -1.64
     lucky
    -1.64
     Lesser
    -1.59
     shrinking
    -1.59
     shortening
    -1.59
     small
    -1.57
    ¿½
    -1.54
     conjugation
    -1.53
     selfish
    -1.52
    POSITIVE LOGITS
    heet
    2.22
    RPC
    1.83
    mith
    1.83
    heets
    1.81
    anne
    1.80
    uit
    1.80
    chitz
    1.80
    urface
    1.76
    dorff
    1.75
    ue
    1.74
    Activations Density 0.142%

    No Known Activations