© 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. 2-RES-SM
    11. 31665
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "know" in various contexts

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    302
    +0.15
    0.8%
    197
    +0.12
    0.7%
    407
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    211
    +0.15
    0.03
    246
    +0.12
    0.04
    358
    +0.11
    0.04
    Negative Logits
    Ģ
    -3.54
    ĭ
    -3.48
    ľ
    -3.40
    ĨĴ
    -3.36
    IJ
    -3.31
    Ń
    -3.28
    Ŀ
    -3.17
    ħ
    -3.13
    į
    -3.07
    Ľ
    -3.05
    POSITIVE LOGITS
     how
    1.63
     why
    1.61
     nothing
    1.56
    ensed
    1.40
     society
    1.40
     naught
    1.39
     SO
    1.39
     sorry
    1.39
     Japanese
    1.37
     commerce
    1.33
    Activations Density 0.155%

    No Known Activations