© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 12393
    Prev
    Next
    INDEX
    Explanations

    instances of humor and laughter 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__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    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
    56
    +0.17
    1.0%
    225
    +0.13
    0.7%
    347
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.17
    0.01
    347
    +0.13
    0.05
    148
    +0.12
    0.01
    Negative Logits
    ļ
    -1.96
    ij
    -1.88
    ·¸
    -1.84
    ĵ
    -1.75
    aliana
    -1.73
    ı
    -1.70
    ħ
    -1.68
     orbit
    -1.61
     publication
    -1.59
    ľĵ
    -1.59
    POSITIVE LOGITS
     aloud
    2.13
     loudly
    2.07
    backed
    1.90
     louder
    1.88
    stuff
    1.84
    rous
    1.83
     hello
    1.80
    stick
    1.79
     loud
    1.78
    ified
    1.74
    Activations Density 0.788%

    No Known Activations