© 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. 4-RES-SM
    11. 13776
    Prev
    Next
    INDEX
    Explanations

    references to judicial courts and legal systems

    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
    78
    +0.16
    1.0%
    255
    +0.14
    0.9%
    469
    +0.09
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    27
    +0.16
    0.01
    237
    +0.14
    0.01
    144
    +0.09
    0.01
    Negative Logits
    aries
    -1.65
    irtual
    -1.38
    ging
    -1.34
     driver
    -1.32
     promot
    -1.31
     caret
    -1.30
     sticks
    -1.29
     promoters
    -1.27
     manufacturers
    -1.26
     drivers
    -1.24
    POSITIVE LOGITS
    ı
    2.41
    Ħ
    2.18
    ²
    2.14
    ķ
    2.12
    ħ
    2.12
    Ļ
    2.06
    Į
    2.06
    ĥ
    2.05
    Ķ
    2.04
    ĥ½
    2.03
    Activations Density 0.014%

    No Known Activations