© 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. 29808
    Prev
    Next
    INDEX
    Explanations

    phrases or concepts associated with wisdom

    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
    369
    +0.17
    1.0%
    410
    +0.16
    0.9%
    376
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    317
    +0.17
    0.01
    410
    +0.16
    0.01
    437
    +0.12
    0.01
    Negative Logits
    upon
    -1.63
     why
    -1.50
     superior
    -1.50
     remorse
    -1.48
    thood
    -1.47
     Negro
    -1.40
    awning
    -1.39
     half
    -1.39
     racial
    -1.34
     coffee
    -1.33
    POSITIVE LOGITS
    £
    2.19
    hered
    2.15
    ģ
    2.05
    Ĥ¬
    2.00
    ĸ
    1.97
    ĻĤ
    1.94
    etable
    1.93
    olver
    1.93
    º
    1.93
    eman
    1.71
    Activations Density 0.016%

    No Known Activations