© 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. 10241
    Prev
    Next
    INDEX
    Explanations

    verbs indicating states of being or existence

    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
    23
    +0.28
    1.6%
    478
    +0.18
    1.1%
    337
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.28
    0.21
    356
    +0.18
    0.14
    247
    +0.14
    0.15
    Negative Logits
    ł
    -5.26
    č↵
    -5.09
    ↵
    -5.09
    -5.09
    č↵č↵
    -5.09
       ↵
    -5.09
    ↵
    -5.09
    <|outofrange|>
    -5.09
    -5.09
    ↵↵
    -5.09
    POSITIVE LOGITS
    rael
    2.12
     worth
    1.91
    abella
    1.86
     always
    1.82
     usual
    1.72
    '));
    1.70
     hers
    1.67
     okay
    1.67
     supposed
    1.66
     definitely
    1.65
    Activations Density 1.582%

    No Known Activations