© 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. 4-RES-SM
    11. 9358
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "mit."

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    21
    +0.14
    0.8%
    30
    +0.13
    0.7%
    218
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    218
    +0.14
    0.02
    437
    +0.13
    0.02
    134
    +0.13
    0.02
    Negative Logits
    plist
    -1.47
    orem
    -1.44
    lections
    -1.44
     convenience
    -1.43
    {$
    -1.41
    example
    -1.40
    LEX
    -1.39
    cases
    -1.36
    AtIndex
    -1.35
    >{
    -1.34
    POSITIVE LOGITS
    ģ
    3.75
    Ļ
    3.60
    Ħ
    3.52
    ĭ
    3.49
    ª
    3.46
    Ĥ
    3.39
    ¾
    3.37
    ¿
    3.36
    ĸ
    3.33
    Į
    3.31
    Activations Density 0.118%

    No Known Activations