© 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. 27588
    Prev
    Next
    INDEX
    Explanations

    instances of the word 'see' and related expectations

    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
    54
    +0.14
    0.8%
    358
    +0.14
    0.8%
    108
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    108
    +0.14
    0.06
    358
    +0.14
    0.05
    492
    +0.13
    0.04
    Negative Logits
    į
    -3.33
    ¥
    -3.28
    ¸
    -3.19
    ĻĤ
    -3.19
    ´
    -3.16
    ī
    -3.14
    ¤
    -3.13
    ģ
    -3.12
    Ń
    -3.11
    IJ
    -3.10
    POSITIVE LOGITS
     how
    1.83
     clearer
    1.75
    iqu
    1.67
     Caption
    1.63
     sight
    1.57
     justice
    1.53
     glimps
    1.49
    dust
    1.48
     canals
    1.46
     flames
    1.46
    Activations Density 0.153%

    No Known Activations