© 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. 9735
    Prev
    Next
    INDEX
    Explanations

    references to specific medical devices and their characteristics

    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
    156
    +0.43
    2.8%
    9
    +0.18
    1.2%
    21
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    9
    +0.43
    0.07
    386
    +0.18
    0.03
    351
    +0.12
    0.08
    Negative Logits
    åĬ¡
    -1.50
     COMPANY
    -1.44
    த
    -1.42
     gratitude
    -1.39
     lessons
    -1.38
    ãĥ¥
    -1.38
    }})$
    -1.34
    }})$.
    -1.33
    веÑĢ
    -1.32
    à®
    -1.30
    POSITIVE LOGITS
    onical
    2.14
    ³
    1.92
    uration
    1.83
    ial
    1.81
    ural
    1.80
    ary
    1.77
    itic
    1.76
    odal
    1.76
    ional
    1.75
    riction
    1.71
    Activations Density 0.611%

    No Known Activations