© 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. 4-RES-SM
    11. 10425
    Prev
    Next
    INDEX
    Explanations

    the brand "Knoll" in various 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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    21
    +0.18
    1.0%
    369
    +0.13
    0.7%
    342
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.18
    0.01
    350
    +0.13
    0.02
    279
    +0.11
    0.02
    Negative Logits
    ged
    -1.66
    duc
    -1.59
    liness
    -1.57
     mention
    -1.50
    ging
    -1.47
    sis
    -1.47
    inary
    -1.46
    lement
    -1.41
    eration
    -1.41
    nement
    -1.41
    POSITIVE LOGITS
    Į
    3.00
    ŀ
    2.76
    ¤
    2.73
    §
    2.66
    ī
    2.65
    Ĺ
    2.61
    ĵ
    2.60
    ij
    2.58
    Ļ
    2.58
    ª
    2.56
    Activations Density 0.146%

    No Known Activations