© 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. 1-RES-SM
    11. 28193
    Prev
    Next
    INDEX
    Explanations

    topics related to infrastructure and its development

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.21
    1.3%
    365
    +0.13
    0.8%
    162
    +0.10
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    162
    +0.21
    0.02
    365
    +0.13
    0.02
    156
    +0.10
    0.02
    Negative Logits
    ĥ½
    -2.14
    §
    -2.10
    Ī
    -2.04
    ĩ
    -1.93
    ¯
    -1.83
    woke
    -1.81
     guilty
    -1.80
    º
    -1.79
    ª
    -1.77
    ı
    -1.76
    POSITIVE LOGITS
    "}](#
    1.90
    frame
    1.77
     surrounds
    1.74
     consists
    1.70
     generates
    1.69
     exists
    1.66
     infrastructure
    1.56
    blocks
    1.54
    isms
    1.49
    building
    1.48
    Activations Density 0.108%

    No Known Activations