© 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. 2-RES-SM
    11. 10160
    Prev
    Next
    INDEX
    Explanations

    references to scientific or medical classifications and codes

    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
    156
    +0.20
    1.2%
    276
    +0.12
    0.7%
    235
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    334
    +0.20
    0.02
    276
    +0.12
    0.01
    486
    +0.12
    0.01
    Negative Logits
     tips
    -1.64
    ĥ½
    -1.57
     other
    -1.56
     sudden
    -1.56
    opause
    -1.55
     else
    -1.54
    cluding
    -1.46
     advice
    -1.44
    cess
    -1.40
    where
    -1.40
    POSITIVE LOGITS
    uple
    1.96
    KN
    1.79
    ialog
    1.75
    Gs
    1.69
    ROM
    1.68
    Ks
    1.66
    elijk
    1.61
    gon
    1.46
    riterion
    1.45
    ORN
    1.41
    Activations Density 0.064%

    No Known Activations