© 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 EvalsExportsAPI 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. 30549
    Prev
    Next
    INDEX
    Explanations

    variations of the word "can" indicating capability or possibility

    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
    48
    +0.19
    1.1%
    23
    +0.13
    0.8%
    159
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    48
    +0.19
    0.03
    183
    +0.13
    0.02
    355
    +0.13
    0.02
    Negative Logits
    »¿
    -4.19
    ¢
    -3.89
    -3.86
       ↵
    -3.86
    č↵
    -3.86
    -3.86
    <|outofrange|>
    -3.86
                ↵
    -3.86
    -3.86
    -3.86
    POSITIVE LOGITS
    't
    2.05
    opy
    1.87
     enumerate
    1.50
    osing
    1.46
     anyone
    1.46
    otec
    1.41
     anybody
    1.39
    --"
    1.39
    tering
    1.38
     mine
    1.37
    Activations Density 0.044%

    No Known Activations