© 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. 24259
    Prev
    Next
    INDEX
    Explanations

    punctuation and sentence-ending stresses

    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
    23
    +0.22
    1.3%
    295
    +0.12
    0.7%
    53
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.22
    0.09
    271
    +0.12
    0.02
    202
    +0.11
    0.05
    Negative Logits
    liga
    -1.83
    urname
    -1.51
    illac
    -1.50
     expires
    -1.50
     Stats
    -1.47
     career
    -1.46
     rankings
    -1.43
    aret
    -1.41
    elines
    -1.38
    auses
    -1.36
    POSITIVE LOGITS
     enough
    1.88
    stood
    1.72
     none
    1.55
    âĢĬ
    1.53
    âĢī
    1.44
     ascribed
    1.43
     plenty
    1.42
    =========================
    1.38
     attributed
    1.38
     (*
    1.36
    Activations Density 0.609%

    No Known Activations