© 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. 2-RES-SM
    11. 9942
    Prev
    Next
    INDEX
    Explanations

    references to coffee and related beverages

    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
    156
    +0.22
    1.3%
    56
    +0.13
    0.8%
    38
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.22
    0.00
    173
    +0.13
    0.03
    328
    +0.11
    0.03
    Negative Logits
    į
    -2.29
    ĺ
    -2.12
    ŀ
    -2.08
    ·
    -2.00
    Ŀ
    -2.00
    ©
    -1.92
    ij
    -1.86
    °
    -1.82
    Ĩ
    -1.73
    Ł
    -1.72
    POSITIVE LOGITS
    starter
    2.12
    shop
    2.00
    holders
    1.90
    houses
    1.90
    cloth
    1.85
    ership
    1.84
    pow
    1.71
     sale
    1.69
    idity
    1.68
     prices
    1.67
    Activations Density 0.311%

    No Known Activations