© 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. 26536
    Prev
    Next
    INDEX
    Explanations

    plurals marked with backticks and other special formatting characters

    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
    478
    +0.20
    1.1%
    410
    +0.19
    1.1%
    221
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    368
    +0.20
    0.02
    50
    +0.19
    0.02
    441
    +0.13
    0.01
    Negative Logits
    »¿
    -1.85
    Ł
    -1.81
    strom
    -1.80
    opf
    -1.72
    icago
    -1.68
    wagen
    -1.65
    ourse
    -1.63
    ocardi
    -1.62
    ygen
    -1.62
    ivals
    -1.60
    POSITIVE LOGITS
     pardon
    1.68
     plenty
    1.61
     sudo
    1.61
     git
    1.55
     eat
    1.54
     scorn
    1.53
     frankly
    1.53
     literally
    1.52
     nowhere
    1.49
     friends
    1.45
    Activations Density 0.113%

    No Known Activations