© 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. 12179
    Prev
    Next
    INDEX
    Explanations

    references to certain individuals and their actions or roles in various contexts

    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
    307
    +0.16
    0.9%
    232
    +0.12
    0.7%
    469
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    478
    +0.16
    0.01
    455
    +0.12
    0.03
    111
    +0.10
    0.05
    Negative Logits
    ¡
    -2.32
    ł
    -2.21
    £
    -2.18
    §
    -2.08
    ľĵ
    -2.04
    Ļª
    -2.04
    ¿
    -2.03
    ¿½
    -2.01
    ©
    -2.01
    ¤
    -2.00
    POSITIVE LOGITS
     E
    4.35
    E
    3.74
    EA
    2.91
     EB
    2.87
     EC
    2.86
    EB
    2.85
     ED
    2.79
     e
    2.78
    EEE
    2.77
     ET
    2.75
    Activations Density 1.788%

    No Known Activations