© 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. 2-RES-SM
    11. 27786
    Prev
    Next
    INDEX
    Explanations

    negative values associated with numerical statements or comparisons

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    369
    +0.15
    0.8%
    358
    +0.13
    0.7%
    125
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.15
    0.03
    461
    +0.13
    0.03
    445
    +0.12
    0.01
    Negative Logits
    imet
    -1.91
     suite
    -1.67
    \'
    -1.49
     tests
    -1.48
    oxin
    -1.47
     attention
    -1.47
     targets
    -1.47
     [@
    -1.47
    alities
    -1.46
    ectomy
    -1.45
    POSITIVE LOGITS
    IJ
    2.62
    ĺ
    2.52
    Ĵ
    2.51
    ¸
    2.26
    ĵ
    2.18
    Ŀ
    2.08
    ¡
    2.07
    ĨĴ
    2.04
    ¥
    2.02
    ·
    2.01
    Activations Density 0.056%

    No Known Activations