© 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. 30013
    Prev
    Next
    INDEX
    Explanations

    mentions of additional or optional content

    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
    +0.13
    0.7%
    376
    +0.13
    0.7%
    129
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.13
    0.02
    323
    +0.13
    0.02
    340
    +0.12
    0.02
    Negative Logits
    ½
    -3.14
    Ĥ¬
    -2.87
    ģ
    -2.80
    Ĵ
    -2.76
    ĩ
    -2.74
    ĥ½
    -2.68
    ī
    -2.64
    ²
    -2.61
    ĨĴ
    -2.54
    Ģ
    -2.54
    POSITIVE LOGITS
     than
    1.78
     reporting
    1.67
     layers
    1.62
    Tick
    1.61
     protections
    1.50
     weeks
    1.50
     charge
    1.48
    ising
    1.47
    ively
    1.44
    Seconds
    1.44
    Activations Density 0.073%

    No Known Activations