© 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. 12593
    Prev
    Next
    INDEX
    Explanations

    references to significant awards and recognitions in film

    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
    366
    +0.13
    0.7%
    85
    +0.12
    0.7%
    429
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    100
    +0.13
    0.09
    226
    +0.12
    0.08
    88
    +0.11
    0.03
    Negative Logits
    ł
    -2.28
    ĻĤ
    -2.03
    ĥ½
    -2.00
    ¢
    -2.00
    £
    -1.94
    ĸ
    -1.94
    ľĵ
    -1.92
    ¼
    -1.91
    »
    -1.87
    ©
    -1.85
    POSITIVE LOGITS
     film
    4.88
     films
    4.80
     Films
    4.79
     filmmakers
    4.68
     cinema
    4.66
     Film
    4.62
     movie
    4.45
     filmm
    4.43
     Cinema
    4.36
     movies
    4.28
    Activations Density 3.089%

    No Known Activations