© 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. 9594
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and their frequency in a text

    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
    478
    +0.20
    1.2%
    23
    +0.16
    0.9%
    203
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.20
    0.09
    136
    +0.16
    -0.00
    23
    +0.15
    0.09
    Negative Logits
    ings
    -2.13
    :`
    -1.84
    ,
    -1.79
    aska
    -1.78
    stown
    -1.75
    !),
    -1.71
    arium
    -1.66
    ,\[
    -1.63
    )!
    -1.62
    -âĤ¬
    -1.58
    POSITIVE LOGITS
    omitempty
    2.06
    unnumbered
    1.94
    ı
    1.53
    000
    1.52
    '"
    1.51
    ]'
    1.47
    ]"
    1.46
    ±
    1.45
    '">
    1.45
     ÑĩÑĤо
    1.44
    Activations Density 0.853%

    No Known Activations