© 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. 31269
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "of" 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
    466
    +0.14
    0.8%
    161
    +0.12
    0.7%
    123
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.14
    0.05
    328
    +0.12
    0.05
    162
    +0.11
    0.05
    Negative Logits
    Ĭ
    -1.78
    ī
    -1.73
    enne
    -1.58
    vier
    -1.58
    ¤
    -1.56
    ";
    -1.51
    Ģ
    -1.50
    Ĩ
    -1.48
    ouin
    -1.42
    ı
    -1.38
    POSITIVE LOGITS
     chances
    1.66
    acity
    1.66
     trouble
    1.60
     luck
    1.57
     beating
    1.56
     ways
    1.55
     excellence
    1.54
     fancy
    1.51
    advant
    1.49
     hassle
    1.48
    Activations Density 0.218%

    No Known Activations