© 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. 31548
    Prev
    Next
    INDEX
    Explanations

    emotional states and attributes related to struggle and resilience

    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
    56
    +0.22
    1.3%
    186
    +0.16
    0.9%
    334
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.22
    0.01
    198
    +0.16
    0.07
    334
    +0.11
    0.09
    Negative Logits
     Catalog
    -1.73
    :&
    -1.68
    abel
    -1.60
     edited
    -1.57
     dated
    -1.56
     updated
    -1.55
    able
    -1.52
     uploaded
    -1.48
     normalized
    -1.48
    UTF
    -1.45
    POSITIVE LOGITS
    ness
    3.29
     spirits
    1.97
     beings
    1.91
     souls
    1.89
    NESS
    1.88
     fury
    1.79
    mare
    1.77
     villagers
    1.77
    liness
    1.71
    ly
    1.68
    Activations Density 1.409%

    No Known Activations