© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 13148
    Prev
    Next
    INDEX
    Explanations

    phrases indicating potential health issues related to smoking and its effects on the body

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    240
    +0.13
    0.7%
    105
    +0.11
    0.6%
    261
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    240
    +0.13
    0.01
    15
    +0.11
    0.06
    289
    +0.11
    0.02
    Negative Logits
     magazines
    -1.62
    @
    -1.61
     calculator
    -1.61
     footage
    -1.58
     lists
    -1.54
     Paper
    -1.45
     Atty
    -1.44
    ↵³³
    -1.43
    cout
    -1.43
     surveys
    -1.42
    POSITIVE LOGITS
     mechanisms
    2.21
     pathophys
    2.09
     mediating
    2.02
     mechanism
    1.96
    ł
    1.92
     pathogenesis
    1.91
     predis
    1.88
    ij
    1.85
     pathways
    1.79
     causative
    1.78
    Activations Density 0.520%

    No Known Activations