© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 4-RES-SM
    11. 12169
    Prev
    Next
    INDEX
    Explanations

    words or terms related to authority and governance

    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
    393
    +0.13
    0.7%
    320
    +0.12
    0.7%
    342
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    320
    +0.13
    0.08
    229
    +0.12
    0.07
    393
    +0.12
    0.02
    Negative Logits
    )$)
    -1.64
    ĻĤ
    -1.50
    ))$,
    -1.45
    .);
    -1.43
     mell
    -1.40
    Awards
    -1.38
    )):
    -1.38
    ¿½
    -1.38
    }$),
    -1.37
    .),
    -1.36
    POSITIVE LOGITS
     be
    2.72
     begin
    2.54
     evolve
    2.53
     propagate
    2.53
     develop
    2.48
     occur
    2.45
     shine
    2.44
     grow
    2.44
     arise
    2.42
     accumulate
    2.41
    Activations Density 0.749%

    No Known Activations