© 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. 1-RES-SM
    11. 29560
    Prev
    Next
    INDEX
    Explanations

    references to modifications or changes in a system

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    2
    +0.14
    0.7%
    253
    +0.12
    0.7%
    498
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    2
    +0.14
    0.02
    253
    +0.12
    0.01
    195
    +0.12
    0.01
    Negative Logits
     same
    -1.62
    effectiveness
    -1.58
     Commonwealth
    -1.57
    coming
    -1.49
     sight
    -1.45
    OPLE
    -1.43
    vert
    -1.41
     newly
    -1.39
     People
    -1.38
     utmost
    -1.37
    POSITIVE LOGITS
    ifies
    2.62
    ifiers
    2.54
    ellers
    2.12
    icum
    2.12
    ifiable
    2.03
    ems
    2.01
    esty
    1.93
    ificate
    1.89
    ific
    1.84
    eller
    1.84
    Activations Density 0.057%

    No Known Activations