© 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. 13921
    Prev
    Next
    INDEX
    Explanations

    references to presidencies and the concept of a president

    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
    463
    +0.12
    0.7%
    369
    +0.11
    0.6%
    437
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    463
    +0.12
    0.02
    181
    +0.11
    0.02
    350
    +0.10
    0.02
    Negative Logits
    Į
    -2.66
    -2.46
    <|outofrange|>
    -2.46
     ↵
    -2.46
     ↵
    -2.46
    ↵
    -2.46
              ↵
    -2.46
    ↵
    -2.46
    ↵
    -2.46
    -2.46
    POSITIVE LOGITS
    byterian
    2.22
    idents
    2.09
    idency
    2.05
    ynaptic
    2.01
    erved
    1.87
    iders
    1.78
    cription
    1.77
    ence
    1.75
    idium
    1.74
    idential
    1.74
    Activations Density 0.115%

    No Known Activations