© 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. 11491
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate judgments or evaluations

    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
    397
    +0.15
    0.9%
    466
    +0.13
    0.8%
    484
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    286
    +0.15
    0.09
    58
    +0.13
    0.09
    466
    +0.12
    0.09
    Negative Logits
    wards
    -1.68
     jointly
    -1.48
     complementary
    -1.40
    .$$
    -1.37
     strept
    -1.31
     together
    -1.29
    $$
    -1.28
     hydrochlor
    -1.27
    :</
    -1.21
     successively
    -1.20
    POSITIVE LOGITS
     nor
    4.43
    nor
    3.72
     sino
    3.34
     Instead
    3.00
     Rather
    2.88
     anymore
    2.85
     slightest
    2.66
     merely
    2.61
     nada
    2.60
     anything
    2.60
    Activations Density 1.296%

    No Known Activations