© 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. 14257
    Prev
    Next
    INDEX
    Explanations

    phrases related to programming and data processing operations

    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
    156
    +0.26
    1.6%
    271
    +0.22
    1.3%
    47
    +0.20
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.26
    0.13
    271
    +0.22
    -0.04
    73
    +0.20
    0.04
    Negative Logits
     deserve
    -1.89
    â
    -1.85
     bedrooms
    -1.49
     likely
    -1.49
    ÂĴ
    -1.49
     cul
    -1.42
     deserved
    -1.39
     trouble
    -1.38
     handic
    -1.36
     families
    -1.35
    POSITIVE LOGITS
    Ĵ
    2.42
    »
    2.35
    Ń
    2.29
    ¶
    2.29
    ĭ
    2.25
    ²
    2.23
    ¯
    2.20
    Ļª
    2.16
    Ĥ
    2.14
    ĻĤ
    2.13
    Activations Density 3.017%

    No Known Activations