© 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. 29665
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to names and places, specifically focusing on geographic and personal identifiers

    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
    156
    +0.26
    1.6%
    369
    +0.18
    1.1%
    410
    +0.17
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.26
    0.10
    369
    +0.18
    0.07
    341
    +0.17
    0.08
    Negative Logits
     motiv
    -1.70
     basics
    -1.63
    actic
    -1.50
     *)
    -1.48
    rapeut
    -1.41
    *\
    -1.40
     necess
    -1.39
    bsd
    -1.38
    )).
    -1.37
     ();
    -1.36
    POSITIVE LOGITS
    ²
    2.54
    ´
    2.53
    »
    2.50
    ³
    2.46
    ¾
    2.44
    ¶
    2.41
    ī
    2.40
    Ń
    2.38
    ¿
    2.37
    Į
    2.35
    Activations Density 0.579%

    No Known Activations