© 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. 9475
    Prev
    Next
    INDEX
    Explanations

    references to specific political figures and their actions

    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
    156
    +0.22
    1.3%
    410
    +0.14
    0.8%
    235
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.22
    0.08
    235
    +0.14
    0.10
    279
    +0.14
    0.09
    Negative Logits
    ĥ½
    -5.77
    <|outofrange|>
    -5.38
    -5.38
    -5.38
    ↵
    -5.38
    č↵
    -5.38
              ↵
    -5.38
    ↵
    -5.38
    ↵  ³³³
    -5.38
     ↵
    -5.38
    POSITIVE LOGITS
     himself
    2.71
     Jr
    2.59
     presidency
    2.10
    's
    2.07
     Presidential
    2.06
     supporters
    1.99
    himself
    1.94
     inaug
    1.90
    ‘
    1.89
    esque
    1.87
    Activations Density 0.992%

    No Known Activations