© 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. 14050
    Prev
    Next
    INDEX
    Explanations

    terms related to events or activities that result in loss or harm

    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
    21
    +0.16
    0.9%
    255
    +0.13
    0.7%
    315
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.16
    0.02
    255
    +0.13
    -0.01
    410
    +0.13
    0.04
    Negative Logits
    s
    -2.05
    sburg
    -1.84
    tet
    -1.83
    resis
    -1.59
    namespace
    -1.53
    going
    -1.52
    same
    -1.46
    hire
    -1.45
    equation
    -1.45
    pta
    -1.44
    POSITIVE LOGITS
    į
    3.91
    ¿
    3.82
    ¾
    3.79
    Į
    3.73
    ĥ
    3.71
    Ļ
    3.63
    Ĩ
    3.49
    Ł
    3.48
    ¥
    3.46
    ĩ
    3.43
    Activations Density 0.586%

    No Known Activations