© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 11824
    Prev
    Next
    INDEX
    Explanations

    references to bathtubs

    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
    21
    +0.28
    1.7%
    406
    +0.13
    0.8%
    503
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.28
    0.03
    218
    +0.13
    0.04
    255
    +0.11
    -0.01
    Negative Logits
    ents
    -1.46
     exception
    -1.32
    light
    -1.31
     Empire
    -1.31
     lesson
    -1.30
    mered
    -1.27
    ous
    -1.27
    plicity
    -1.25
     respect
    -1.24
     double
    -1.23
    POSITIVE LOGITS
    ĥ
    2.57
    §
    2.46
    ħ
    2.25
    Ģ
    2.24
    ©
    2.12
    Ļ
    2.09
    ¨
    2.09
    Į
    2.00
    ¿
    1.92
    ·¸
    1.92
    Activations Density 0.407%

    No Known Activations