© 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. 2-RES-SM
    11. 11108
    Prev
    Next
    INDEX
    Explanations

    terms related to denial or rejection

    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
    369
    +0.23
    1.3%
    494
    +0.14
    0.8%
    90
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    494
    +0.23
    0.02
    59
    +0.14
    0.02
    253
    +0.13
    0.02
    Negative Logits
    "}](#
    -1.58
     restricted
    -1.51
    ķ
    -1.43
    ères
    -1.38
     wheelchair
    -1.37
    Ĺ
    -1.35
     rid
    -1.34
    Ķ
    -1.34
     spleen
    -1.33
    ession
    -1.32
    POSITIVE LOGITS
    holm
    1.98
    omination
    1.91
    unciation
    1.82
    ich
    1.78
    forcement
    1.78
    unci
    1.76
    iners
    1.74
    joy
    1.73
    hardt
    1.67
    ounce
    1.67
    Activations Density 0.146%

    No Known Activations