© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 32220
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and quotation marks

    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
    478
    +0.15
    0.8%
    111
    +0.14
    0.8%
    466
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    466
    +0.15
    0.01
    142
    +0.14
    0.01
    162
    +0.13
    0.01
    Negative Logits
    gif
    -1.52
    gium
    -1.45
     wore
    -1.43
    imes
    -1.42
    gens
    -1.40
     dollars
    -1.39
    born
    -1.38
    phone
    -1.38
    LINE
    -1.36
     quarters
    -1.36
    POSITIVE LOGITS
    Ĥ
    1.44
    Tak
    1.41
    æĪij们
    1.40
    opal
    1.39
    ahu
    1.36
    Theorem
    1.35
     Neural
    1.33
     Presidente
    1.33
    TRODUCTION
    1.31
    oretical
    1.30
    Activations Density 0.031%

    No Known Activations