© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 30895
    Prev
    Next
    INDEX
    Explanations

    references to musical performance and emotional expression in songs

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    204
    +0.14
    0.8%
    121
    +0.13
    0.7%
    293
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    204
    +0.14
    0.20
    56
    +0.13
    -0.08
    362
    +0.13
    0.18
    Negative Logits
    ĻĤ
    -2.04
    ĸ´
    -1.92
    Ł
    -1.80
    ¿½
    -1.75
    Ģ
    -1.73
    Ĩ
    -1.69
    ĩ
    -1.69
    ij
    -1.68
    ī
    -1.67
    Ĥ
    -1.63
    POSITIVE LOGITS
     nost
    1.99
     backdrop
    1.87
     thrill
    1.86
     imagin
    1.82
     atmosphere
    1.78
     delight
    1.77
     cue
    1.70
     milieu
    1.65
     aroma
    1.63
     scent
    1.61
    Activations Density 10.374%

    No Known Activations