© 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. 1-RES-SM
    11. 26606
    Prev
    Next
    INDEX
    Explanations

    references to safety and the concept of being safe 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/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
    86
    +0.12
    0.7%
    394
    +0.12
    0.7%
    79
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    86
    +0.12
    0.03
    394
    +0.12
    0.03
    506
    +0.12
    0.02
    Negative Logits
    ĨĴ
    -2.61
    Ķ
    -2.53
    ©
    -2.31
    ¯
    -2.23
    ĵ
    -2.20
    £
    -2.19
    Ĺ
    -2.17
    ij
    -2.07
    Ĩ
    -1.92
    Į
    -1.91
    POSITIVE LOGITS
    guards
    2.31
    holder
    1.82
    gu
    1.73
    keeping
    1.66
    holders
    1.64
    iences
    1.60
     microwave
    1.48
     bases
    1.48
     ACC
    1.43
     batteries
    1.42
    Activations Density 0.113%

    No Known Activations