© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 14266
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "pet" or related terms associated with domesticated animals

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-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.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    369
    +0.16
    0.9%
    156
    +0.15
    0.9%
    497
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    18
    +0.16
    0.02
    497
    +0.15
    0.01
    474
    +0.13
    0.01
    Negative Logits
    µ
    -1.81
     abrupt
    -1.73
     fois
    -1.63
     wake
    -1.59
     favor
    -1.58
     vez
    -1.55
    orems
    -1.52
    versely
    -1.50
    visibility
    -1.50
    Ŀ
    -1.49
    POSITIVE LOGITS
    roleum
    2.65
    rified
    2.03
    finder
    1.97
    ty
    1.94
    tico
    1.85
    rophys
    1.83
    rels
    1.78
    terson
    1.78
    ious
    1.76
    chio
    1.71
    Activations Density 0.136%

    No Known Activations