© 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. 1-RES-SM
    11. 27154
    Prev
    Next
    INDEX
    Explanations

    specific foreign language characters or symbols

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    410
    +0.16
    0.9%
    203
    +0.14
    0.8%
    111
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.16
    0.02
    203
    +0.14
    0.01
    86
    +0.13
    0.01
    Negative Logits
    asone
    -1.65
    ctomy
    -1.62
    izations
    -1.50
    cedes
    -1.46
    AndroidRuntime
    -1.45
    Biography
    -1.41
    quisition
    -1.40
    ectomy
    -1.39
    elijk
    -1.39
    responding
    -1.39
    POSITIVE LOGITS
    ¥
    1.70
    ¬
    1.61
    ī
    1.54
    Ĩ
    1.52
     ticks
    1.51
    ¿½
    1.50
    ı
    1.46
    ¤
    1.43
    Ĭ
    1.42
    ŀ
    1.40
    Activations Density 0.069%

    No Known Activations