© 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. 9797
    Prev
    Next
    INDEX
    Explanations

    references to specific data points or metrics in a text

    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
    9
    +0.13
    0.7%
    65
    +0.13
    0.7%
    4
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    456
    +0.13
    0.03
    9
    +0.13
    0.02
    273
    +0.11
    0.01
    Negative Logits
    arner
    -1.68
    antic
    -1.42
    ise
    -1.42
    mare
    -1.41
     youth
    -1.37
     engine
    -1.37
    atile
    -1.37
     Maker
    -1.34
     engines
    -1.33
     Jet
    -1.33
    POSITIVE LOGITS
    ![](
    3.41
    »
    2.73
    Ļ
    2.50
    ¨
    2.45
    ######
    2.43
    ²
    2.42
    ¤
    2.40
    ĭ
    2.39
    ĸ
    2.34
    §
    2.32
    Activations Density 0.103%

    No Known Activations