© 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. 26961
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter 'G' as part of a larger set of ratings or classifications

    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
    369
    +0.25
    1.5%
    412
    +0.17
    1.0%
    156
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.25
    0.04
    410
    +0.17
    0.05
    412
    +0.16
    0.05
    Negative Logits
    aintiff
    -1.82
    istant
    -1.71
    coming
    -1.61
    ction
    -1.60
    cludes
    -1.51
    correct
    -1.49
    cedes
    -1.48
     time
    -1.45
    ications
    -1.45
    gent
    -1.45
    POSITIVE LOGITS
    Ĵ
    2.45
    ª
    2.43
    Ľ
    2.34
    ĵ
    2.10
    ķ
    2.07
    £
    2.03
    ½
    1.97
    ŀ
    1.95
    Ĭ
    1.92
    ´
    1.90
    Activations Density 0.123%

    No Known Activations