© 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. 4-RES-SM
    11. 12508
    Prev
    Next
    INDEX
    Explanations

    the word "along" at various frequencies in the 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/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
    412
    +0.13
    0.8%
    466
    +0.12
    0.7%
    325
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    412
    +0.13
    0.04
    373
    +0.12
    0.04
    328
    +0.12
    0.03
    Negative Logits
    Ĩ
    -1.75
    ogs
    -1.70
    ously
    -1.47
    AFFIRMED
    -1.37
    ized
    -1.33
    olated
    -1.31
     reporters
    -1.31
     maker
    -1.31
    ocortic
    -1.30
    oflu
    -1.26
    POSITIVE LOGITS
    with
    2.03
    side
    1.98
    áĢº
    1.91
    shore
    1.89
    lining
    1.85
     with
    1.84
    vall
    1.73
    leftrightarrow
    1.68
    coming
    1.67
    irection
    1.67
    Activations Density 0.174%

    No Known Activations