© 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. 4-RES-SM
    11. 11757
    Prev
    Next
    INDEX
    Explanations

    terms related to default settings and configurations

    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
    309
    +0.14
    0.8%
    231
    +0.14
    0.8%
    307
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    157
    +0.14
    0.03
    170
    +0.14
    0.02
    129
    +0.14
    0.01
    Negative Logits
    thems
    -2.05
    athi
    -1.72
     itself
    -1.66
    alus
    -1.56
    osecond
    -1.54
    igue
    -1.46
    nai
    -1.45
    arynge
    -1.43
    ivism
    -1.40
    aka
    -1.39
    POSITIVE LOGITS
    ı
    2.69
    ¥
    2.55
     settings
    2.54
    ´
    2.31
    Ļª
    2.29
    Ŀ
    2.27
     values
    2.27
    Ĺ
    2.25
    Ľ
    2.20
     constructor
    2.19
    Activations Density 0.324%

    No Known Activations