© 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. 12729
    Prev
    Next
    INDEX
    Explanations

    repetitive phrases or clauses that begin with "that"

    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
    182
    +0.15
    0.8%
    123
    +0.13
    0.7%
    478
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    182
    +0.15
    0.14
    88
    +0.13
    0.12
    123
    +0.13
    0.12
    Negative Logits
    ring
    -1.50
    eyer
    -1.48
    enthal
    -1.38
     quarters
    -1.38
    °
    -1.35
    umbered
    -1.33
    ees
    -1.32
    ű
    -1.28
    red
    -1.27
    ];
    -1.27
    POSITIVE LOGITS
     same
    1.96
     there
    1.54
     although
    1.47
    áĢ¯
    1.46
    áĢº
    1.41
     despite
    1.39
     they
    1.37
    opy
    1.37
    orect
    1.33
    opsy
    1.31
    Activations Density 0.500%

    No Known Activations