© 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. 10747
    Prev
    Next
    INDEX
    Explanations

    references to television shows and media content

    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
    123
    +0.14
    0.8%
    148
    +0.12
    0.7%
    271
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    123
    +0.14
    0.04
    148
    +0.12
    0.03
    194
    +0.11
    0.02
    Negative Logits
    ges
    -1.79
    ITION
    -1.62
    ect
    -1.54
    ership
    -1.51
     concerned
    -1.50
    ECT
    -1.50
    han
    -1.49
    ection
    -1.46
    hood
    -1.44
    ging
    -1.41
    POSITIVE LOGITS
    ŀ
    3.16
    ©
    3.10
    ĻĤ
    3.07
    Ĵ
    3.02
    ij
    2.92
    IJ
    2.81
    ĸ
    2.81
    ĨĴ
    2.76
    enario
    2.75
    Ń
    2.72
    Activations Density 0.064%

    No Known Activations