© 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. 10607
    Prev
    Next
    INDEX
    Explanations

    repeated mentions of the word "the" 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
    +0.13
    0.8%
    123
    +0.13
    0.8%
    57
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    123
    +0.13
    0.02
    225
    +0.13
    0.02
    302
    +0.10
    0.02
    Negative Logits
     oneself
    -1.61
     ourselves
    -1.57
     itself
    -1.47
    ilate
    -1.36
    ional
    -1.36
    hin
    -1.34
    naire
    -1.31
    ment
    -1.29
     anybody
    -1.28
     themselves
    -1.26
    POSITIVE LOGITS
    ¿
    2.58
    ľ
    2.54
    º
    2.53
    ·
    2.52
    IJ
    2.47
    ŀ
    2.44
    Ľ
    2.43
    ¶
    2.40
    ¸
    2.39
    ĨĴ
    2.39
    Activations Density 0.031%

    No Known Activations