© 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. 25804
    Prev
    Next
    INDEX
    Explanations

    instances of the word "when"

    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
    191
    +0.13
    0.7%
    364
    +0.13
    0.7%
    318
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    174
    +0.13
    0.01
    112
    +0.13
    0.01
    364
    +0.11
    0.01
    Negative Logits
    ¦
    -1.89
    ·¸
    -1.87
    «
    -1.86
    Ĭ
    -1.79
    ¿½
    -1.71
    ľ
    -1.57
    µ
    -1.57
    ĸ´
    -1.53
    »¿
    -1.53
    ¬
    -1.49
    POSITIVE LOGITS
    actory
    1.64
    vere
    1.64
     2030
    1.57
    ball
    1.56
    stage
    1.55
    aker
    1.51
    quarter
    1.48
     underway
    1.48
     basket
    1.46
    onial
    1.46
    Activations Density 0.062%

    No Known Activations