© 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 EvalsExportsAPI 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. 26124
    Prev
    Next
    INDEX
    Explanations

    conjunctions suggesting connections or relationships between ideas

    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
    494
    +0.23
    1.4%
    478
    +0.17
    1.0%
    209
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    494
    +0.23
    0.18
    209
    +0.17
    0.13
    258
    +0.13
    0.10
    Negative Logits
    ĸ
    -2.16
    Ľ
    -1.87
    £
    -1.79
    ł
    -1.78
    Ļ
    -1.74
    Ģ
    -1.71
    ģ
    -1.66
    xe
    -1.62
    µ
    -1.58
    ½
    -1.56
    POSITIVE LOGITS
    rogens
    1.85
    naments
    1.65
     then
    1.60
     finals
    1.54
     vice
    1.53
     Lemmon
    1.50
     forth
    1.48
     remanded
    1.48
     its
    1.44
    alus
    1.44
    Activations Density 0.797%

    No Known Activations