© 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. 1-RES-SM
    11. 25278
    Prev
    Next
    INDEX
    Explanations

    references to life or living conditions

    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
    200
    +0.14
    0.8%
    118
    +0.13
    0.7%
    406
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    200
    +0.14
    0.02
    348
    +0.13
    0.02
    406
    +0.12
    0.02
    Negative Logits
    beit
    -1.70
    arthy
    -1.69
    arnings
    -1.62
    hift
    -1.56
    vasive
    -1.54
    mit
    -1.43
    hib
    -1.42
    ://
    -1.42
    lain
    -1.41
    pher
    -1.40
    POSITIVE LOGITS
    ģ
    2.98
    Ļ
    2.85
    Ĭ
    2.80
    ¢
    2.77
    Ģ
    2.76
    º
    2.67
    Ħ
    2.65
    ĭ
    2.59
    ·¸
    2.58
    ī
    2.57
    Activations Density 0.137%

    No Known Activations