© 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. 29379
    Prev
    Next
    INDEX
    Explanations

    the word "by" in various contexts

    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
    478
    +0.18
    1.0%
    23
    +0.17
    1.0%
    69
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    26
    +0.18
    0.06
    46
    +0.17
    0.06
    204
    +0.13
    0.08
    Negative Logits
    ^](#
    -1.68
    vier
    -1.66
     served
    -1.56
    heimer
    -1.45
    legged
    -1.43
    ledge
    -1.38
    documentclass
    -1.37
     argues
    -1.37
    ivist
    -1.34
    âĢIJ
    -1.34
    POSITIVE LOGITS
     controlling
    1.61
     removing
    1.52
    ogen
    1.51
     descent
    1.51
     default
    1.49
    ↵
    1.46
     autom
    1.46
    ť
    1.46
     separating
    1.45
     brute
    1.43
    Activations Density 0.295%

    No Known Activations