© 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. 1-RES-SM
    11. 25790
    Prev
    Next
    INDEX
    Explanations

    variations of the word "exchange" in different 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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    376
    +0.13
    0.7%
    484
    +0.13
    0.7%
    203
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    225
    +0.13
    0.02
    221
    +0.13
    0.02
    347
    +0.12
    0.02
    Negative Logits
    ly
    -1.71
    rely
    -1.63
     something
    -1.62
    kered
    -1.58
     sleep
    -1.48
     rise
    -1.46
     accent
    -1.45
    ests
    -1.42
     Adventure
    -1.42
     mention
    -1.41
    POSITIVE LOGITS
    baum
    1.80
    iens
    1.75
    ariate
    1.70
    haus
    1.69
    dor
    1.68
    erior
    1.68
     variables
    1.64
    able
    1.60
    breaker
    1.60
    ¢
    1.58
    Activations Density 0.098%

    No Known Activations