© 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. 25357
    Prev
    Next
    INDEX
    Explanations

    function definitions and procedural constructs in programming code

    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
    96
    +0.13
    0.8%
    397
    +0.12
    0.7%
    484
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    96
    +0.13
    0.03
    484
    +0.12
    0.03
    67
    +0.11
    0.02
    Negative Logits
    ¤
    -2.19
    ¸
    -2.15
    »
    -2.01
    ĩ
    -2.01
    Ļª
    -1.94
    ĨĴ
    -1.91
    ī
    -1.84
     Channel
    -1.84
    ´
    -1.80
    ħ
    -1.77
    POSITIVE LOGITS
    nell
    1.67
    urer
    1.65
    eries
    1.63
     apprent
    1.52
     resume
    1.49
     coff
    1.46
    iones
    1.45
    erals
    1.43
    ional
    1.40
    ÃŃvel
    1.40
    Activations Density 0.037%

    No Known Activations