© 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. 4-RES-SM
    11. 10407
    Prev
    Next
    INDEX
    Explanations

    legal references and terms related to court procedures and rules

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    495
    +0.14
    0.8%
    206
    +0.14
    0.8%
    78
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    206
    +0.14
    0.06
    495
    +0.14
    0.06
    230
    +0.13
    0.03
    Negative Logits
     impression
    -1.74
     \$
    -1.43
     compliment
    -1.43
     Europeans
    -1.39
     wheel
    -1.38
    ]{.
    -1.38
     odds
    -1.38
     chance
    -1.35
     £
    -1.31
     output
    -1.30
    POSITIVE LOGITS
    (
    2.19
    (@
    1.82
    (&
    1.73
     AFFIRMED
    1.61
     reserved
    1.57
     (“
    1.53
    icles
    1.51
    aga
    1.51
     unconstitutional
    1.50
    amble
    1.47
    Activations Density 0.374%

    No Known Activations