© 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. 12788
    Prev
    Next
    INDEX
    Explanations

    instances of the word "from."

    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
    494
    +0.12
    0.7%
    203
    +0.12
    0.7%
    68
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.12
    0.01
    391
    +0.12
    0.02
    68
    +0.12
    0.02
    Negative Logits
     behalf
    -1.65
    chaft
    -1.56
     meantime
    -1.48
    rapeut
    -1.47
     warrant
    -1.46
     present
    -1.41
     consequence
    -1.39
    ctor
    -1.39
     expecting
    -1.36
    boards
    -1.35
    POSITIVE LOGITS
    ´
    2.01
    ±
    2.00
    ¨
    1.98
    ®
    1.90
    etheus
    1.84
    °
    1.78
    ongs
    1.77
    ©
    1.76
    ·
    1.76
    uscular
    1.76
    Activations Density 0.112%

    No Known Activations