© 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. 10291
    Prev
    Next
    INDEX
    Explanations

    prepositions and conjunctions that indicate relationships between elements in a sentence

    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
    419
    +0.15
    0.9%
    263
    +0.15
    0.9%
    198
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.15
    0.03
    69
    +0.15
    0.03
    300
    +0.13
    0.03
    Negative Logits
    mare
    -1.80
    iscus
    -1.61
    argin
    -1.57
    eanor
    -1.54
    inces
    -1.53
     nost
    -1.51
     spines
    -1.50
     faç
    -1.47
    atura
    -1.46
    pires
    -1.45
    POSITIVE LOGITS
     treating
    1.85
     diagnosing
    1.66
     Reporting
    1.63
     dealing
    1.59
     ascertain
    1.48
     implementing
    1.42
     handling
    1.41
     obtaining
    1.41
     genotyping
    1.40
     manufacture
    1.40
    Activations Density 0.144%

    No Known Activations