© 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. 31160
    Prev
    Next
    INDEX
    Explanations

    copyright-related terms and phrases

    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
    453
    +0.15
    0.8%
    298
    +0.14
    0.8%
    410
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    453
    +0.15
    0.04
    486
    +0.14
    0.02
    298
    +0.13
    0.03
    Negative Logits
    weight
    -1.52
     grounded
    -1.52
    pie
    -1.50
     cooker
    -1.49
    Ł
    -1.48
    guard
    -1.45
    [/
    -1.40
    plug
    -1.40
     cater
    -1.40
    !--
    -1.36
    POSITIVE LOGITS
    ÃŃc
    1.76
    icul
    1.65
    liography
    1.46
     ]$
    1.43
    izar
    1.41
     HOLDERS
    1.40
    ération
    1.39
     us
    1.38
    ções
    1.33
    veau
    1.32
    Activations Density 0.022%

    No Known Activations