© 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. 30848
    Prev
    Next
    INDEX
    Explanations

    specific types of financial and legal terminology related to transactions and agreements

    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
    459
    +0.14
    0.8%
    262
    +0.13
    0.8%
    425
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    425
    +0.14
    0.10
    57
    +0.13
    0.09
    66
    +0.13
    0.10
    Negative Logits
    »
    -2.88
    ķ
    -2.54
    ©
    -2.40
    Ĥ¬
    -2.37
    ¹
    -2.35
    ¯
    -2.31
    ±
    -2.30
    ²
    -2.28
    ľĵ
    -2.24
    Ĥ
    -2.21
    POSITIVE LOGITS
    aurus
    2.00
     Association
    1.91
    wagen
    1.73
     Federation
    1.72
     Championships
    1.71
     Organization
    1.68
     Award
    1.68
     Authority
    1.67
    clerosis
    1.67
     Against
    1.66
    Activations Density 1.284%

    No Known Activations