© 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. 1-RES-SM
    11. 26057
    Prev
    Next
    INDEX
    Explanations

    references to television show episodes and their impact on viewership

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    +0.26
    1.5%
    507
    +0.12
    0.7%
    177
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    507
    +0.26
    0.02
    10
    +0.12
    0.02
    177
    +0.11
    0.02
    Negative Logits
    ĥ½
    -2.67
    ĺ
    -1.99
    §
    -1.95
    ½
    -1.93
    Ń
    -1.90
    ľĵ
    -1.81
    Ŀ
    -1.80
    ¢
    -1.77
    Ĵ
    -1.70
    ↵↵
    -1.70
    POSITIVE LOGITS
     highlights
    1.96
     episode
    1.94
     aired
    1.94
    manship
    1.81
     excerpt
    1.77
     bursts
    1.72
     footage
    1.65
     scene
    1.64
    nikov
    1.62
    isode
    1.62
    Activations Density 0.169%

    No Known Activations