© 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. 28507
    Prev
    Next
    INDEX
    Explanations

    phrases involving personal connections and social interactions

    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
    186
    +0.15
    0.9%
    302
    +0.13
    0.8%
    91
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    91
    +0.15
    0.20
    455
    +0.13
    0.21
    95
    +0.13
    0.17
    Negative Logits
    ľĵ
    -2.95
    ģ
    -2.93
    ¿
    -2.86
    ŀ
    -2.70
    Ŀ
    -2.67
    °
    -2.66
    ¥
    -2.65
    ļ
    -2.59
    Ļª
    -2.58
    ¤
    -2.54
    POSITIVE LOGITS
    obox
    1.50
    matics
    1.37
    oa
    1.36
    ERTY
    1.32
     DOI
    1.32
    OTA
    1.30
    bio
    1.29
    jcmm
    1.26
    hib
    1.26
    à³
    1.26
    Activations Density 8.519%

    No Known Activations