© 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. 13504
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "Introduction" and related thematic headings in a document

    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
    111
    +0.17
    0.9%
    79
    +0.12
    0.7%
    157
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    79
    +0.17
    0.01
    131
    +0.12
    0.01
    345
    +0.12
    0.01
    Negative Logits
    ."_
    -1.76
    _."
    -1.73
     penn
    -1.58
    rame
    -1.53
     heads
    -1.52
     hers
    -1.51
    iless
    -1.49
    ...](
    -1.46
     bear
    -1.45
    onica
    -1.45
    POSITIVE LOGITS
    åĬ¡
    1.59
     INTRODUCTION
    1.47
    ault
    1.43
     Introduction
    1.42
    ility
    1.41
     cytokines
    1.40
    aur
    1.37
     donc
    1.33
    emis
    1.32
     uncertain
    1.31
    Activations Density 0.023%

    No Known Activations