© 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. 25200
    Prev
    Next
    INDEX
    Explanations

    terms related to mechanical or structural components in technical 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
    507
    +0.13
    0.8%
    384
    +0.13
    0.8%
    147
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.13
    0.08
    57
    +0.13
    0.03
    87
    +0.12
    0.03
    Negative Logits
     silence
    -1.65
    ships
    -1.55
     jer
    -1.51
    rain
    -1.49
     bc
    -1.44
    rais
    -1.43
     bans
    -1.41
    ner
    -1.40
     ih
    -1.35
     fan
    -1.32
    POSITIVE LOGITS
    ères
    1.57
    ŀ
    1.53
    ieur
    1.52
    case
    1.47
    amed
    1.44
     substituted
    1.38
    stated
    1.37
    ère
    1.37
    valued
    1.34
     Iter
    1.33
    Activations Density 1.645%

    No Known Activations