© 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. 27347
    Prev
    Next
    INDEX
    Explanations

    instances of proper nouns and related terms

    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
    419
    +0.19
    1.1%
    63
    +0.14
    0.8%
    12
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    104
    +0.19
    0.06
    228
    +0.14
    0.07
    79
    +0.13
    0.10
    Negative Logits
    ĻĤ
    -3.33
    Ļª
    -2.85
    ľĵ
    -2.75
    ¿½
    -2.55
    Ĺ
    -2.30
    ŀ
    -2.29
    ¼
    -2.21
    ´
    -2.12
    ®
    -2.08
    º
    -2.06
    POSITIVE LOGITS
    :`
    1.71
     athletes
    1.39
     Jury
    1.38
     athlete
    1.34
     degrees
    1.33
     occasion
    1.31
     prescribe
    1.30
     Exercise
    1.29
     appealed
    1.29
     able
    1.28
    Activations Density 4.469%

    No Known Activations