© 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. 4-RES-SM
    11. 15167
    Prev
    Next
    INDEX
    Explanations

    phrases related to structural and economic challenges

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    374
    +0.13
    0.8%
    242
    +0.13
    0.7%
    215
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    215
    +0.13
    0.07
    186
    +0.13
    0.06
    401
    +0.12
    0.03
    Negative Logits
    č↵č↵
    -4.92
    ↵↵
    -4.92
    -4.92
    ↵  Âł
    -4.92
         ↵
    -4.92
    ↵
    -4.92
    -4.92
     ↵
    -4.92
     ↵
    -4.92
    -4.92
    POSITIVE LOGITS
     inherent
    2.31
    iciency
    1.82
     constraints
    1.77
    ality
    1.76
    ulence
    1.68
     prevailing
    1.67
     availability
    1.64
     limitations
    1.62
    unction
    1.53
    istics
    1.51
    Activations Density 0.801%

    No Known Activations