© 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. 2-RES-SM
    11. 28227
    Prev
    Next
    INDEX
    Explanations

    mentions of art and artistic practices

    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
    156
    +0.19
    1.1%
    204
    +0.15
    0.9%
    31
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    204
    +0.19
    0.13
    33
    +0.15
    0.12
    31
    +0.15
    0.14
    Negative Logits
    Īĺ
    -3.85
    Ī
    -3.65
    »
    -3.58
    ĥ½
    -3.57
    -3.55
    <|outofrange|>
    -3.55
    ↵
    -3.55
    č↵
    -3.55
    ↵
    -3.55
    -3.55
    POSITIVE LOGITS
     artist
    2.91
     artists
    2.73
     artistic
    2.50
     painting
    2.47
     painter
    2.33
     inspiration
    2.32
     photography
    2.29
     sculpture
    2.26
     Artist
    2.20
     artwork
    2.16
    Activations Density 4.178%

    No Known Activations