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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 28807
    Prev
    Next
    INDEX
    Explanations

    references to geographic locations, specifically regions that include the term "North."

    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
    +0.17
    0.9%
    369
    +0.14
    0.8%
    373
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    503
    +0.17
    0.03
    373
    +0.14
    0.02
    38
    +0.11
    0.02
    Negative Logits
    ĥ½
    -1.76
     cref
    -1.69
    zz
    -1.64
    doibase
    -1.63
    ľĵ
    -1.60
    nlm
    -1.60
    ção
    -1.53
    ilent
    -1.47
    ιÏĤ
    -1.39
    zione
    -1.37
    POSITIVE LOGITS
    umberland
    2.45
    cliffe
    2.37
    wick
    2.08
    field
    1.89
    view
    1.84
    wood
    1.80
    ampton
    1.76
     Shore
    1.72
     East
    1.71
    west
    1.71
    Activations Density 0.061%

    No Known Activations