© 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. 10498
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Tom" and related commands or references in code

    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
    287
    +0.16
    0.9%
    265
    +0.12
    0.7%
    269
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.16
    0.03
    265
    +0.12
    -0.00
    204
    +0.12
    0.02
    Negative Logits
     Jensen
    -1.83
    ENCE
    -1.74
    ensen
    -1.65
    CRIPT
    -1.59
    ental
    -1.52
    amiento
    -1.51
    varrho
    -1.49
    EMENT
    -1.48
    iviral
    -1.46
    VEGF
    -1.46
    POSITIVE LOGITS
     Cairo
    2.60
     cat
    2.60
     Cat
    2.57
     Egypt
    2.44
     Che
    2.38
     cats
    2.32
     Alexandria
    2.29
     Cle
    2.25
     Egyptian
    2.21
    Cat
    2.12
    Activations Density 1.217%

    No Known Activations