© 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. 2-RES-SM
    11. 13882
    Prev
    Next
    INDEX
    Explanations

    phrases related to sitting or resting actions

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    383
    +0.14
    0.8%
    56
    +0.14
    0.8%
    255
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.14
    -0.01
    16
    +0.14
    0.08
    406
    +0.13
    0.09
    Negative Logits
    iest
    -1.69
    far
    -1.61
    iator
    -1.60
    code
    -1.58
    icillin
    -1.50
    largest
    -1.45
    agic
    -1.44
    иб
    -1.43
    à°
    -1.43
    from
    -1.39
    POSITIVE LOGITS
    escence
    1.90
    rientation
    1.72
     Discussion
    1.70
     pillow
    1.64
     recl
    1.62
     chair
    1.60
     cush
    1.58
     outdoors
    1.56
     chairs
    1.56
    Argued
    1.55
    Activations Density 2.894%

    No Known Activations