© 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. 2-MLP-SM
    11. 23370
    Prev
    Next
    INDEX
    Explanations

    references to girls and female characters in various contexts

    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/2-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.2.hook_mlp_out
    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
    42
    +0.18
    1.0%
    104
    +0.14
    0.8%
    23
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    461
    +0.18
    0.07
    162
    +0.14
    0.07
    42
    +0.12
    0.08
    Negative Logits
    ipes
    -1.64
     venues
    -1.49
     mont
    -1.43
     que
    -1.42
    icum
    -1.41
    áĥĶ
    -1.39
     parce
    -1.35
    que
    -1.35
    quel
    -1.33
     toile
    -1.33
    POSITIVE LOGITS
    ):=\
    1.66
    AML
    1.64
    cke
    1.60
    ":"
    1.60
    bibr
    1.56
     conjecture
    1.53
    hä
    1.51
    ":
    1.51
     further
    1.50
    inned
    1.50
    Activations Density 0.693%

    No Known Activations