© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 30701
    Prev
    Next
    INDEX
    Explanations

    references to regulatory actions and compliance measures in governance 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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    186
    +0.12
    0.7%
    181
    +0.11
    0.6%
    358
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    247
    +0.12
    0.15
    24
    +0.11
    0.13
    72
    +0.10
    0.09
    Negative Logits
    ĻĤ
    -2.61
    ĨĴ
    -2.48
    Ļª
    -2.42
    Ĺ
    -2.33
    Ģ
    -2.33
    IJ
    -2.20
    Ĵ
    -2.08
    ĥ½
    -2.08
    ĭ
    -2.08
    ĩ
    -2.07
    POSITIVE LOGITS
     appoint
    1.91
     payroll
    1.86
     vested
    1.79
     fiduciary
    1.72
     lobbying
    1.72
     publ
    1.71
     theirs
    1.67
     issuing
    1.66
     hiring
    1.65
     reimb
    1.64
    Activations Density 4.598%

    No Known Activations