© 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. 28721
    Prev
    Next
    INDEX
    Explanations

    phrases related to business services and customer support

    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
    421
    +0.12
    0.7%
    216
    +0.12
    0.7%
    396
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.12
    0.15
    0
    +0.12
    0.14
    55
    +0.11
    0.13
    Negative Logits
    athi
    -1.58
    shoot
    -1.58
    ($_
    -1.53
    ('/
    -1.48
    ugu
    -1.47
    onica
    -1.46
    aver
    -1.46
     spectrometer
    -1.45
     <-
    -1.45
    (__
    -1.41
    POSITIVE LOGITS
    liness
    1.48
     Century
    1.46
    ķ
    1.37
    eness
    1.37
    ncia
    1.31
    =========================
    1.25
     citizenship
    1.24
    IOR
    1.23
    agement
    1.23
     Ju
    1.20
    Activations Density 0.640%

    No Known Activations