© 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. 4-RES-SM
    11. 13333
    Prev
    Next
    INDEX
    Explanations

    the presence of the infinitive form of verbs and their variations

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    428
    +0.14
    0.8%
    35
    +0.12
    0.6%
    434
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    434
    +0.14
    0.06
    393
    +0.12
    0.03
    69
    +0.11
    0.06
    Negative Logits
    nes
    -1.94
    liness
    -1.68
     owed
    -1.67
    ication
    -1.67
    osto
    -1.65
    like
    -1.65
     likes
    -1.63
    ulla
    -1.58
    ãĤī
    -1.58
    esque
    -1.58
    POSITIVE LOGITS
     determine
    2.73
     evaluate
    2.54
     assess
    2.49
     minimize
    2.34
     ensure
    2.33
     elucidate
    2.32
     characterize
    2.32
     describe
    2.31
     clarify
    2.31
     investigate
    2.30
    Activations Density 0.278%

    No Known Activations