© 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. 10694
    Prev
    Next
    INDEX
    Explanations

    guidance on safety practices and precautions in various situations

    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
    +0.21
    1.2%
    156
    +0.18
    1.1%
    306
    +0.17
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.21
    0.14
    67
    +0.18
    -0.01
    327
    +0.17
    0.03
    Negative Logits
     talent
    -1.64
     ont
    -1.59
     roy
    -1.59
     transformed
    -1.53
     triumph
    -1.49
     discovery
    -1.48
     identities
    -1.46
     homology
    -1.46
     computational
    -1.45
     stage
    -1.44
    POSITIVE LOGITS
     precautions
    2.96
    ĨĴ
    2.79
     precaution
    2.57
    autions
    2.48
     preventive
    2.41
    ¯
    2.40
     Avoid
    2.40
     caution
    2.40
    ¼
    2.35
     hygiene
    2.33
    Activations Density 3.420%

    No Known Activations