© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 11197
    Prev
    Next
    INDEX
    Explanations

    recurring references to frequency or the act of doing something repeatedly

    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
    343
    +0.11
    0.6%
    23
    +0.11
    0.6%
    359
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    363
    +0.11
    0.02
    242
    +0.11
    0.02
    65
    +0.10
    0.02
    Negative Logits
    Ħ
    -1.90
    ļ
    -1.61
     partnership
    -1.57
    İ
    -1.56
    IJ
    -1.56
     damn
    -1.55
     potential
    -1.54
     basic
    -1.52
    tle
    -1.51
    !”
    -1.50
    POSITIVE LOGITS
     cited
    1.67
    etin
    1.51
     abused
    1.50
     encountered
    1.49
    positions
    1.48
     underestimated
    1.48
    chro
    1.46
    vist
    1.46
    ETHOD
    1.44
     menstru
    1.44
    Activations Density 0.090%

    No Known Activations