© 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. 30207
    Prev
    Next
    INDEX
    Explanations

    instances of the word "watch" and related terms

    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
    376
    +0.14
    0.8%
    115
    +0.14
    0.8%
    507
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    11
    +0.14
    0.03
    52
    +0.14
    0.03
    235
    +0.12
    0.02
    Negative Logits
    ]>
    -1.62
    \!\!
    -1.56
    ){#
    -1.52
    \!
    -1.51
    ANE
    -1.49
     suppose
    -1.44
    ANC
    -1.40
    summary
    -1.38
    RA
    -1.37
    EL
    -1.35
    POSITIVE LOGITS
    able
    1.99
    ership
    1.95
    fulness
    1.95
    dog
    1.90
    ably
    1.81
     videos
    1.69
    ké
    1.67
    avi
    1.64
    ful
    1.60
    ied
    1.59
    Activations Density 0.614%

    No Known Activations