© 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. 10947
    Prev
    Next
    INDEX
    Explanations

    questions about code functionality and debugging help requests

    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
    348
    +0.15
    0.8%
    39
    +0.12
    0.6%
    231
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    390
    +0.15
    0.05
    476
    +0.12
    0.04
    39
    +0.11
    0.02
    Negative Logits
     ([**
    -1.73
    .^[@
    -1.64
    **](#
    -1.61
    ,^[@
    -1.59
    ]{.
    -1.59
    *](#
    -1.48
    icul
    -1.43
    ↵Č
    -1.41
    "}](#
    -1.35
    ITC
    -1.33
    POSITIVE LOGITS
    fiddle
    2.21
     ERROR
    2.08
     wont
    2.02
     dont
    1.98
     jquery
    1.90
    documentclass
    1.83
     doesnt
    1.82
     Cannot
    1.80
     please
    1.77
     alert
    1.75
    Activations Density 0.227%

    No Known Activations