© 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. 2-RES-SM
    11. 10244
    Prev
    Next
    INDEX
    Explanations

    references to plaintiffs in legal 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
    410
    +0.16
    0.9%
    34
    +0.12
    0.6%
    159
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    155
    +0.16
    0.02
    297
    +0.12
    0.02
    410
    +0.11
    0.02
    Negative Logits
     same
    -1.67
    yme
    -1.63
    indows
    -1.63
    esium
    -1.47
    ognition
    -1.45
    msgstr
    -1.44
    eeper
    -1.43
    oop
    -1.43
     locations
    -1.39
    izzard
    -1.38
    POSITIVE LOGITS
    ĸ´
    2.10
    ocular
    1.93
    ¨
    1.84
     alleges
    1.83
    ·
    1.64
    ī
    1.63
    Į
    1.60
    ²
    1.58
    Ķ
    1.57
     alleged
    1.56
    Activations Density 0.034%

    No Known Activations