© 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 EvalsExportsAPI 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. 26580
    Prev
    Next
    INDEX
    Explanations

    assertions and claims made in a scientific or analytical context

    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
    23
    +0.32
    1.9%
    376
    +0.10
    0.6%
    478
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    376
    +0.32
    0.04
    23
    +0.10
    0.07
    511
    +0.10
    0.04
    Negative Logits
    .
    -1.90
    #,
    -1.86
    dimensional
    -1.80
    ournal
    -1.76
    urations
    -1.76
    ĨĴ
    -1.73
    #.
    -1.69
    /~
    -1.68
     –
    -1.68
    #:
    -1.61
    POSITIVE LOGITS
     grateful
    1.98
     our
    1.68
     mutual
    1.58
     deference
    1.56
     ours
    1.54
     gratitude
    1.53
     antis
    1.49
     sincere
    1.48
    beit
    1.45
     mine
    1.44
    Activations Density 0.686%

    No Known Activations