© 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 EvalsExportsinterp-engineNEWAPI 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. 31378
    Prev
    Next
    INDEX
    Explanations

    phrases indicating authorship or attribution

    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
    63
    +0.17
    1.0%
    478
    +0.12
    0.7%
    216
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    407
    +0.17
    0.05
    123
    +0.12
    0.05
    81
    +0.12
    0.04
    Negative Logits
    ifice
    -1.65
     yourselves
    -1.57
     pages
    -1.52
    aders
    -1.50
     advisers
    -1.49
     heels
    -1.45
    ipers
    -1.43
     next
    -1.42
    iers
    -1.40
    into
    -1.40
    POSITIVE LOGITS
    ĥ½
    2.42
    Į
    2.31
    Ĺ
    2.20
    ĨĴ
    2.19
    ¶
    2.17
    ©
    2.10
    ľĵ
    2.00
    Ļª
    1.99
    £
    1.93
    ¨
    1.91
    Activations Density 0.448%

    No Known Activations