© 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. 4-RES-SM
    11. 10294
    Prev
    Next
    INDEX
    Explanations

    instances of the word "at" in varying 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/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
    307
    +0.23
    1.3%
    306
    +0.18
    1.0%
    69
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.23
    0.10
    429
    +0.18
    0.06
    150
    +0.12
    0.08
    Negative Logits
    Ģ
    -2.08
    ¢
    -2.08
    ner
    -1.92
    ette
    -1.86
     thereof
    -1.85
    ettes
    -1.84
    ¾
    -1.81
    ķ
    -1.80
    upt
    -1.69
     syndrome
    -1.63
    POSITIVE LOGITS
     least
    2.81
    mosp
    2.26
    least
    2.14
     midnight
    1.98
     noon
    1.94
    rium
    1.89
     night
    1.82
     dawn
    1.81
     ease
    1.79
    las
    1.78
    Activations Density 0.262%

    No Known Activations