© 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. 12610
    Prev
    Next
    INDEX
    Explanations

    terms and concepts related to security and safety issues

    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
    313
    +0.13
    0.7%
    470
    +0.11
    0.6%
    426
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    189
    +0.13
    0.05
    418
    +0.11
    0.02
    373
    +0.11
    0.04
    Negative Logits
    urn
    -1.76
    leton
    -1.70
    coming
    -1.67
    tery
    -1.57
    birth
    -1.56
    gg
    -1.55
    <|outofrange|>
    -1.53
    ↵
    -1.53
    -1.53
    -1.53
    POSITIVE LOGITS
     concerns
    2.51
     precautions
    2.51
     issues
    2.42
     considerations
    2.27
     measures
    2.22
     protocols
    2.21
    guards
    2.17
     mechanisms
    2.14
     reasons
    2.12
     precaution
    2.10
    Activations Density 0.344%

    No Known Activations