© 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 EvalsExportsAPI 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. 27503
    Prev
    Next
    INDEX
    Explanations

    opening and closing curly braces in code

    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
    478
    +0.13
    0.7%
    38
    +0.11
    0.6%
    100
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    38
    +0.13
    0.07
    113
    +0.11
    0.06
    175
    +0.11
    0.07
    Negative Logits
    ĻĤ
    -2.13
    Ł
    -2.12
    ¾
    -1.95
    »¿
    -1.93
    ¹
    -1.81
    ij
    -1.78
    ·
    -1.77
    ¿
    -1.73
    Īĺ
    -1.71
    ł
    -1.67
    POSITIVE LOGITS
    ñor
    1.47
     inject
    1.43
    -------------------------------------------------------
    1.40
    race
    1.38
    rogen
    1.37
    xit
    1.34
    amssymb
    1.32
    simpl
    1.31
     neut
    1.30
    LEX
    1.30
    Activations Density 0.160%

    No Known Activations