© 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. 10693
    Prev
    Next
    INDEX
    Explanations

    Spanish words and phrases, particularly those related to health and well-being

    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
    156
    +0.12
    0.7%
    21
    +0.12
    0.7%
    410
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    64
    +0.12
    0.03
    134
    +0.12
    0.03
    410
    +0.11
    0.03
    Negative Logits
    iginally
    -1.92
     possible
    -1.67
    ings
    -1.63
    INGS
    -1.52
     alike
    -1.51
    aday
    -1.48
     others
    -1.48
     products
    -1.47
    ighter
    -1.46
     positions
    -1.46
    POSITIVE LOGITS
    Ī
    2.77
    ģ
    2.77
    ĩ
    2.76
    Ħ
    2.70
    IJ
    2.69
    ¥
    2.61
    §
    2.57
    ¤
    2.52
    ·
    2.52
    ħ
    2.50
    Activations Density 0.144%

    No Known Activations