© 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. 4-RES-SM
    11. 14048
    Prev
    Next
    INDEX
    Explanations

    occurrences of various types of liquid or waste, especially in relation to environmental contexts

    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
    262
    +0.14
    0.7%
    429
    +0.13
    0.7%
    374
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    429
    +0.14
    0.06
    356
    +0.13
    0.11
    262
    +0.11
    0.05
    Negative Logits
    iew
    -1.76
    ÑģÑĤва
    -1.58
    helf
    -1.48
    iking
    -1.46
    izo
    -1.43
    Ñĭ
    -1.42
    oof
    -1.42
    plist
    -1.42
    nament
    -1.42
    matically
    -1.42
    POSITIVE LOGITS
    Ĥ
    3.09
    IJ
    2.81
     energy
    2.79
     nutrients
    2.67
    Ģ
    2.64
    Ĵ
    2.62
    ĵ
    2.53
     water
    2.51
    Ķ
    2.45
    İ
    2.44
    Activations Density 1.635%

    No Known Activations