© 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. 13673
    Prev
    Next
    INDEX
    Explanations

    proper nouns, particularly names of people and places

    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
    293
    +0.15
    0.8%
    263
    +0.15
    0.8%
    441
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    441
    +0.15
    -0.00
    410
    +0.15
    0.12
    474
    +0.13
    0.06
    Negative Logits
    »¿
    -1.80
     deficits
    -1.66
    Ĥ¬
    -1.59
    ĥ½
    -1.59
    achy
    -1.56
    Ĺ
    -1.55
     defe
    -1.54
    Ļª
    -1.49
     nests
    -1.49
     defaults
    -1.47
    POSITIVE LOGITS
    ainer
    1.73
    æ°ı
    1.44
     cref
    1.43
    reference
    1.39
     author
    1.38
    _)
    1.37
     authors
    1.37
     inspiration
    1.36
    igner
    1.33
     CONTRIBUTORS
    1.31
    Activations Density 3.495%

    No Known Activations