© 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. 9984
    Prev
    Next
    INDEX
    Explanations

    terms related to human rights and their violations

    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
    345
    +0.14
    0.8%
    235
    +0.12
    0.6%
    34
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    422
    +0.14
    0.03
    345
    +0.12
    0.03
    235
    +0.11
    0.03
    Negative Logits
    quo
    -2.15
    agher
    -1.58
    boards
    -1.55
    psin
    -1.48
    doors
    -1.46
    pee
    -1.44
    ede
    -1.43
    ptions
    -1.43
    ching
    -1.41
     quo
    -1.41
    POSITIVE LOGITS
     beings
    2.58
    oid
    2.47
    itarian
    2.15
     rights
    2.07
     readable
    2.07
    rights
    2.04
    istic
    1.98
    ist
    1.96
    oids
    1.95
    ITY
    1.86
    Activations Density 0.130%

    No Known Activations