© 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. 28681
    Prev
    Next
    INDEX
    Explanations

    references to rivers and waterways

    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
    256
    +0.13
    0.7%
    71
    +0.12
    0.7%
    507
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.13
    0.02
    143
    +0.12
    0.02
    71
    +0.12
    0.02
    Negative Logits
     respect
    -1.60
     vacc
    -1.60
    EVER
    -1.53
     hands
    -1.51
     NSA
    -1.44
     regard
    -1.44
    athing
    -1.44
     obese
    -1.43
     examiner
    -1.42
    opsy
    -1.41
    POSITIVE LOGITS
    boat
    2.42
    dale
    2.21
    boats
    2.14
    bank
    2.12
     Basin
    2.04
     Plate
    2.00
    ways
    1.97
    pool
    1.93
     Thames
    1.92
    banks
    1.91
    Activations Density 0.109%

    No Known Activations