© 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. 28491
    Prev
    Next
    INDEX
    Explanations

    references to human rights and the qualities of being human

    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
    +0.15
    0.8%
    345
    +0.13
    0.7%
    201
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    345
    +0.15
    0.03
    446
    +0.13
    0.02
    201
    +0.12
    0.02
    Negative Logits
     fois
    -1.73
    quo
    -1.73
    gg
    -1.67
    operative
    -1.62
    cember
    -1.45
    hib
    -1.44
     addition
    -1.41
     commercial
    -1.38
    urope
    -1.38
    olini
    -1.38
    POSITIVE LOGITS
     beings
    1.99
    itarian
    1.96
    oid
    1.90
     immunodeficiency
    1.77
    itat
    1.61
    ities
    1.59
    oids
    1.59
    ist
    1.58
    itar
    1.56
    rights
    1.55
    Activations Density 0.102%

    No Known Activations