© 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. 14396
    Prev
    Next
    INDEX
    Explanations

    legal and procedural terminology related to regulations and compliance

    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
    156
    +0.44
    2.9%
    307
    +0.30
    2.0%
    306
    +0.16
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.44
    0.04
    386
    +0.30
    -0.00
    350
    +0.16
    0.06
    Negative Logits
    ]>
    -1.45
     beings
    -1.25
    ).](
    -1.24
     waves
    -1.21
     institution
    -1.19
    chaft
    -1.19
    ].
    -1.18
    kay
    -1.14
     robots
    -1.14
     assign
    -1.13
    POSITIVE LOGITS
    imum
    1.59
     misses
    1.37
    ably
    1.35
    ichia
    1.32
    volatile
    1.31
     expired
    1.29
    fiction
    1.24
    imal
    1.24
    rile
    1.22
    irubin
    1.21
    Activations Density 2.478%

    No Known Activations