© 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. 1-RES-SM
    11. 28850
    Prev
    Next
    INDEX
    Explanations

    quantitative information about serum levels and biomarkers in medical contexts

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    305
    +0.12
    0.7%
    120
    +0.12
    0.7%
    283
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    283
    +0.12
    0.03
    275
    +0.12
    0.03
    301
    +0.11
    0.02
    Negative Logits
    dule
    -1.58
     devised
    -1.55
     Proceedings
    -1.47
     redes
    -1.46
     purs
    -1.42
    die
    -1.42
     proceedings
    -1.42
     proposes
    -1.40
    oder
    -1.39
     Zend
    -1.38
    POSITIVE LOGITS
    ethyl
    1.79
    Ļ
    1.75
    estine
    1.64
    etic
    1.57
     screening
    1.51
    Ī
    1.47
    iverse
    1.43
    uced
    1.42
     âĪ¼
    1.41
     detection
    1.41
    Activations Density 0.217%

    No Known Activations