© 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. 30640
    Prev
    Next
    INDEX
    Explanations

    instances of the word "calculate" and its variations 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/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
    156
    +0.31
    1.9%
    376
    +0.21
    1.3%
    115
    +0.14
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    257
    +0.31
    0.01
    182
    +0.21
    0.01
    213
    +0.14
    0.01
    Negative Logits
    ĥ½
    -2.00
    ¯
    -1.88
    »
    -1.81
    į
    -1.76
    OH
    -1.75
    ½
    -1.71
    ³
    -1.70
    ĭ
    -1.70
    Ĵ
    -1.63
    ¼
    -1.61
    POSITIVE LOGITS
    ement
    1.54
     accuracy
    1.41
    ably
    1.37
    able
    1.36
    cient
    1.35
    left
    1.35
     incorrect
    1.35
    icted
    1.35
     accurate
    1.35
    ulent
    1.34
    Activations Density 0.007%

    No Known Activations