© 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. 1-RES-SM
    11. 30148
    Prev
    Next
    INDEX
    Explanations

    the word "quote" and its variations in different contexts

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    5
    +0.14
    0.8%
    108
    +0.13
    0.8%
    252
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    352
    +0.14
    0.03
    503
    +0.13
    0.01
    260
    +0.11
    0.03
    Negative Logits
    ij
    -2.75
    Ł
    -2.72
    ·
    -2.71
    »
    -2.53
    Ĥ
    -2.53
    ĥ½
    -2.51
    ¨
    -2.50
    ľĵ
    -2.50
    ¶
    -2.45
    ĺ
    -2.42
    POSITIVE LOGITS
    eries
    2.20
    chn
    2.06
    ery
    1.99
    ÅĽci
    1.98
    ÅĽ
    1.91
    hematic
    1.84
    ÅĽÄĩ
    1.82
    heet
    1.75
    ño
    1.70
    /#
    1.61
    Activations Density 0.273%

    No Known Activations