© 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. 25751
    Prev
    Next
    INDEX
    Explanations

    references to critique and commentary on social and political issues

    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
    +0.25
    1.4%
    23
    +0.19
    1.1%
    369
    +0.19
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    365
    +0.25
    0.13
    156
    +0.19
    0.15
    71
    +0.19
    0.15
    Negative Logits
    vier
    -1.72
     Arts
    -1.66
     Testament
    -1.66
     skills
    -1.53
    lege
    -1.52
     Roth
    -1.46
    thood
    -1.44
    #,
    -1.39
     Wine
    -1.35
     Skills
    -1.35
    POSITIVE LOGITS
     disappro
    1.61
     underestimate
    1.59
     overest
    1.55
     mistakes
    1.54
    shut
    1.50
    wash
    1.49
     authorities
    1.47
     spokesman
    1.45
     observers
    1.45
     mistake
    1.40
    Activations Density 2.918%

    No Known Activations