© 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. 27737
    Prev
    Next
    INDEX
    Explanations

    mentions of law firms or associated organizations

    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
    156
    +0.12
    0.7%
    32
    +0.12
    0.6%
    85
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    32
    +0.12
    0.02
    85
    +0.12
    0.02
    472
    +0.11
    0.02
    Negative Logits
    )](#
    -1.65
     charset
    -1.50
    ©
    -1.50
    ĸ
    -1.43
     correct
    -1.41
     County
    -1.41
    ÄŁ
    -1.39
    CRIPT
    -1.38
    verted
    -1.38
     DISTRICT
    -1.36
    POSITIVE LOGITS
    ament
    2.59
    ness
    2.27
    amental
    2.08
    notes
    2.01
    edly
    1.96
    ship
    1.94
    ees
    1.91
    fellow
    1.72
    liness
    1.71
    oster
    1.68
    Activations Density 0.127%

    No Known Activations