© 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. MLP Post
    9. 0-MLP-SM
    11. 29845
    Prev
    Next
    INDEX
    Explanations

    code structure or organization in programming

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    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
    136
    +0.13
    0.7%
    384
    +0.12
    0.6%
    127
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    48
    +0.13
    0.13
    322
    +0.12
    0.12
    35
    +0.11
    0.11
    Negative Logits
    §
    -1.62
    Ĩ
    -1.55
     attorneys
    -1.48
     minority
    -1.39
    noreply
    -1.38
     affiliate
    -1.36
    ¤
    -1.36
    ķ
    -1.32
    venth
    -1.31
     troop
    -1.30
    POSITIVE LOGITS
    ophyll
    1.82
    washed
    1.50
    OID
    1.45
    life
    1.45
    opus
    1.44
    arc
    1.38
    Simplify
    1.35
    optera
    1.35
     oxidase
    1.31
    ache
    1.31
    Activations Density 0.246%

    No Known Activations