© 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. 12820
    Prev
    Next
    INDEX
    Explanations

    geographical names and terms related to specific places or regions

    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
    21
    +0.29
    1.7%
    500
    +0.11
    0.6%
    477
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    477
    +0.29
    0.07
    86
    +0.11
    0.05
    134
    +0.10
    0.07
    Negative Logits
    ·¸
    -1.97
    ĨĴ
    -1.82
    ¿½
    -1.82
      ↵
    -1.71
    Ļª
    -1.71
    ↵
    -1.70
    -1.70
    ↵
    -1.70
    -1.70
    <|outofrange|>
    -1.70
    POSITIVE LOGITS
     Pradesh
    1.99
    uru
    1.87
    aan
    1.76
    arta
    1.75
    mand
    1.74
     County
    1.67
    ara
    1.66
    appa
    1.65
    ahoma
    1.65
    iya
    1.63
    Activations Density 0.403%

    No Known Activations