© 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. 1-RES-SM
    11. 25110
    Prev
    Next
    INDEX
    Explanations

    references to copyright and ownership rights

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    466
    +0.12
    0.7%
    42
    +0.12
    0.7%
    458
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    65
    +0.12
    0.02
    418
    +0.12
    0.01
    271
    +0.10
    0.00
    Negative Logits
    phrine
    -1.64
    apine
    -1.47
     wrong
    -1.38
     "!"
    -1.33
     case
    -1.33
    formin
    -1.33
     depressive
    -1.32
     brains
    -1.29
     kid
    -1.29
     logarithm
    -1.29
    POSITIVE LOGITS
    igned
    1.95
    ume
    1.60
     Reserved
    1.58
    ue
    1.53
    overed
    1.50
    ù
    1.48
    Åij
    1.44
    ég
    1.43
    ists
    1.41
     extends
    1.40
    Activations Density 0.154%

    No Known Activations