© 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. 1-RES-SM
    11. 26423
    Prev
    Next
    INDEX
    Explanations

    pronouns related to self

    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
    147
    +0.14
    0.8%
    372
    +0.13
    0.7%
    273
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    372
    +0.14
    0.01
    363
    +0.13
    0.01
    338
    +0.12
    0.01
    Negative Logits
    enen
    -1.84
    bent
    -1.64
     glances
    -1.64
    ect
    -1.60
    teen
    -1.56
    ed
    -1.53
    bart
    -1.53
     glance
    -1.50
    heed
    -1.46
    tar
    -1.45
    POSITIVE LOGITS
    §
    3.67
    Ĭ
    3.27
    Į
    3.10
    Ń
    3.06
    ¹
    2.95
    IJ
    2.92
    ½
    2.89
    ¿½
    2.87
    ¸
    2.84
    ¶
    2.82
    Activations Density 0.001%

    No Known Activations