© 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. 24426
    Prev
    Next
    INDEX
    Explanations

    HTML or code structure elements related to formatting

    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
    81
    +0.13
    0.7%
    430
    +0.12
    0.7%
    274
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.13
    0.02
    228
    +0.12
    0.02
    245
    +0.12
    0.02
    Negative Logits
    omitempty
    -1.38
    ODING
    -1.37
     contracting
    -1.36
     afore
    -1.35
     first
    -1.35
    YS
    -1.35
    .")
    -1.34
     done
    -1.32
     appellee
    -1.31
    .");
    -1.31
    POSITIVE LOGITS
    room
    1.76
    ority
    1.72
    ning
    1.66
    ica
    1.61
    dimen
    1.61
    µ
    1.56
    cuts
    1.51
    ®
    1.48
    azi
    1.48
    rooms
    1.44
    Activations Density 0.205%

    No Known Activations