© 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. 26662
    Prev
    Next
    INDEX
    Explanations

    technical terms and definitions related to programming and data structures

    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
    302
    +0.21
    1.2%
    141
    +0.13
    0.7%
    291
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    302
    +0.21
    0.13
    141
    +0.13
    0.10
    170
    +0.12
    0.12
    Negative Logits
    gradle
    -1.54
    ricanes
    -1.53
    абоÑĤ
    -1.51
     Sauce
    -1.50
    áĥĶ
    -1.47
    sson
    -1.47
    eous
    -1.43
    dez
    -1.43
    ricane
    -1.43
    аÑģÑĤ
    -1.42
    POSITIVE LOGITS
    ^](#
    1.66
     includes
    1.56
     value
    1.44
    icates
    1.43
     dated
    1.42
     entit
    1.42
     fee
    1.42
    )$-
    1.42
     lifetime
    1.41
     correspondence
    1.40
    Activations Density 7.860%

    No Known Activations