© 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. 22843
    Prev
    Next
    INDEX
    Explanations

    programming-related terms, particularly those associated with memory allocation and object initialization

    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
    200
    +0.22
    1.2%
    74
    +0.13
    0.7%
    491
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    200
    +0.22
    0.08
    74
    +0.13
    0.03
    312
    +0.11
    0.11
    Negative Logits
    .”
    -1.74
     away
    -1.73
    ctic
    -1.72
    essee
    -1.66
    ”),
    -1.66
    úblic
    -1.63
    ”)
    -1.61
    ,”
    -1.60
    .”).
    -1.60
    ctica
    -1.56
    POSITIVE LOGITS
    ±
    2.55
    ¸
    2.27
    ©
    2.25
    «
    2.13
    ³
    2.06
    ½
    2.03
    ¥
    1.99
    Īĺ
    1.95
    ķ
    1.93
    ¢
    1.93
    Activations Density 2.067%

    No Known Activations