© 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. 26423
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate explicitness or direct statements

    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
    +0.24
    1.4%
    376
    +0.22
    1.3%
    438
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    199
    +0.24
    0.02
    455
    +0.22
    0.01
    438
    +0.15
    0.01
    Negative Logits
    ène
    -2.21
    »¿
    -1.88
    ĨĴ
    -1.82
    inks
    -1.75
    ħ
    -1.73
    abe
    -1.72
    ©
    -1.71
    ĵ
    -1.64
    ribe
    -1.60
    »
    -1.59
    POSITIVE LOGITS
    arial
    1.48
     disclaim
    1.48
     killing
    1.41
    caption
    1.40
    vphantom
    1.37
     Parenthood
    1.36
    Conclusion
    1.31
     punish
    1.26
     recession
    1.24
    dict
    1.22
    Activations Density 3.134%

    No Known Activations