© 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. 26139
    Prev
    Next
    INDEX
    Explanations

    instances of the word "act" in various forms or 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
    352
    +0.13
    0.7%
    388
    +0.12
    0.7%
    496
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    352
    +0.13
    0.02
    43
    +0.12
    0.02
    28
    +0.12
    0.01
    Negative Logits
     mistrial
    -1.50
    acetic
    -1.45
     sudden
    -1.38
     ')
    -1.37
     Cochrane
    -1.35
     shame
    -1.34
     guys
    -1.33
    ĨĴ
    -1.31
     MAL
    -1.31
     polyclonal
    -1.29
    POSITIVE LOGITS
    ivities
    3.41
    uary
    2.66
    ivism
    2.27
    uated
    2.26
    uate
    2.22
    ivated
    2.20
    uates
    2.10
    uating
    2.06
    ives
    2.05
    uality
    1.86
    Activations Density 0.096%

    No Known Activations