© 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. 9353
    Prev
    Next
    INDEX
    Explanations

    numerical values or data, particularly in relation to specific formatting or logical structures

    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
    234
    +0.13
    0.7%
    322
    +0.12
    0.7%
    11
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.13
    -0.01
    136
    +0.12
    -0.01
    101
    +0.12
    0.01
    Negative Logits
    _;
    -1.72
    ·
    -1.65
    Ļª
    -1.59
     _(
    -1.58
    /_
    -1.53
     ultr
    -1.52
     **(
    -1.50
    ,^[@
    -1.49
    »¿
    -1.47
     _
    -1.46
    POSITIVE LOGITS
     rounded
    2.48
    0000
    2.29
    000000
    2.03
    0000000
    2.00
    Round
    1.84
    00000
    1.75
    000001
    1.75
    00000000
    1.73
     Round
    1.69
    UTC
    1.67
    Activations Density 0.096%

    No Known Activations