© 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. 23794
    Prev
    Next
    INDEX
    Explanations

    quantitative descriptors indicating amounts or comparisons

    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
    156
    +0.21
    1.2%
    199
    +0.13
    0.7%
    376
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.21
    0.07
    199
    +0.13
    0.10
    362
    +0.12
    0.11
    Negative Logits
    ¿
    -1.93
    Ĩ
    -1.92
    Ģ
    -1.87
    ¼
    -1.78
    »
    -1.76
    Ī
    -1.74
    ł
    -1.73
    §
    -1.65
    Ĥ
    -1.63
    ´
    -1.62
    POSITIVE LOGITS
     previous
    1.76
     ordinary
    1.64
     outgoing
    1.54
     spot
    1.45
     noct
    1.43
     daytime
    1.42
    >]
    1.39
    atur
    1.38
     const
    1.37
    herin
    1.37
    Activations Density 1.329%

    No Known Activations