© 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. 25465
    Prev
    Next
    INDEX
    Explanations

    technical details related to software versions and file paths

    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
    307
    +0.12
    0.7%
    461
    +0.11
    0.6%
    169
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.12
    0.06
    461
    +0.11
    0.05
    408
    +0.11
    0.06
    Negative Logits
     petitioners
    -1.66
     finance
    -1.53
    imer
    -1.48
    esters
    -1.46
     petitioner
    -1.46
     recap
    -1.44
     germ
    -1.39
     cha
    -1.37
     suit
    -1.37
    §
    -1.35
    POSITIVE LOGITS
    inx
    1.88
    ITERATION
    1.80
    oreal
    1.62
    ARRAY
    1.57
    packages
    1.56
    IDTH
    1.54
    googleapis
    1.54
    cule
    1.54
    <=
    1.51
    ]['
    1.49
    Activations Density 1.342%

    No Known Activations