© 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. 1-RES-SM
    11. 25568
    Prev
    Next
    INDEX
    Explanations

    the presence of the letter "M" 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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    +0.27
    1.6%
    81
    +0.17
    1.0%
    196
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.27
    0.04
    81
    +0.17
    0.04
    423
    +0.12
    0.04
    Negative Logits
    tered
    -1.93
     answers
    -1.76
     answer
    -1.73
    quot
    -1.71
    ked
    -1.69
     complete
    -1.55
     reply
    -1.52
    dered
    -1.50
    king
    -1.50
    effectiveness
    -1.44
    POSITIVE LOGITS
    ı
    2.58
    ĨĴ
    2.37
    į
    2.26
    ĺ
    2.25
    Ĵ
    2.19
    Į
    2.07
    ·
    2.06
    ĭ
    2.04
    ĸ
    1.95
    ij
    1.95
    Activations Density 0.136%

    No Known Activations