© 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. 14610
    Prev
    Next
    INDEX
    Explanations

    numerical values, particularly related to versioning and software updates

    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
    177
    +0.14
    0.8%
    441
    +0.14
    0.8%
    509
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    177
    +0.14
    0.04
    252
    +0.14
    0.02
    509
    +0.13
    0.04
    Negative Logits
    ¿½
    -2.90
    ²
    -2.89
    ¬
    -2.86
    Ĭ
    -2.80
    µ
    -2.78
    ·¸
    -2.75
    ·
    -2.70
    ĸ
    -2.67
    §
    -2.66
    ī
    -2.65
    POSITIVE LOGITS
    +.
    2.37
    ubuntu
    2.30
    GHz
    2.20
    +,
    2.13
     GHz
    2.06
    rc
    2.04
     installed
    1.92
    itrile
    1.84
    cm
    1.81
     Version
    1.71
    Activations Density 0.148%

    No Known Activations