© 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. 4-RES-SM
    11. 10619
    Prev
    Next
    INDEX
    Explanations

    references to ownership or familial relationships

    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
    156
    +0.27
    1.6%
    307
    +0.26
    1.5%
    306
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    137
    +0.27
    0.11
    10
    +0.26
    0.10
    145
    +0.14
    0.07
    Negative Logits
    edly
    -1.77
     thereof
    -1.67
    çļĦ
    -1.56
    ãģ¦ãģĦãģ¾ãģĻ
    -1.52
    acs
    -1.51
    thems
    -1.50
    iei
    -1.48
    ãĥ¼ãĥĪ
    -1.46
     its
    -1.45
    imits
    -1.44
    POSITIVE LOGITS
     goal
    2.42
     motto
    2.32
    Ń
    2.32
     aim
    2.25
    ¨
    2.21
     biggest
    2.20
    ·¸
    2.18
     livelihood
    2.09
    ĥ
    2.08
     motivation
    2.06
    Activations Density 0.911%

    No Known Activations