© 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. 10800
    Prev
    Next
    INDEX
    Explanations

    references to medical conditions and their treatments

    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
    +0.35
    2.1%
    306
    +0.25
    1.5%
    333
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.35
    0.11
    306
    +0.25
    0.06
    447
    +0.12
    0.07
    Negative Logits
    tagged
    -1.54
     offence
    -1.42
     puzzles
    -1.38
     vibe
    -1.38
     Snap
    -1.37
     humidity
    -1.32
     gameplay
    -1.31
     (`
    -1.30
     rain
    -1.30
     unto
    -1.27
    POSITIVE LOGITS
    ¯
    4.88
    ĥ½
    4.59
    ĨĴ
    4.47
    ²
    4.19
    ľĵ
    4.18
    Į
    4.14
    ŀ
    4.13
    ĩ
    4.11
    ³
    4.10
    ī
    4.09
    Activations Density 0.898%

    No Known Activations