© 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. 13111
    Prev
    Next
    INDEX
    Explanations

    references to strength or strong attributes in various contexts

    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
    198
    +0.14
    0.8%
    119
    +0.14
    0.8%
    114
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    119
    +0.14
    0.03
    198
    +0.14
    0.03
    362
    +0.12
    0.03
    Negative Logits
    urai
    -1.78
    usepackage
    -1.67
     photographed
    -1.67
    itian
    -1.63
    iska
    -1.55
     destination
    -1.55
     '')
    -1.47
    TRODUCTION
    -1.46
    abel
    -1.44
    AndroidRuntime
    -1.42
    POSITIVE LOGITS
     enough
    2.67
    ness
    2.62
    holds
    2.10
    enough
    2.05
    ened
    1.96
    eners
    1.92
    est
    1.89
    ening
    1.82
    NESS
    1.80
    mouth
    1.78
    Activations Density 0.202%

    No Known Activations