© 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. 13245
    Prev
    Next
    INDEX
    Explanations

    topics related to economic activities and structures

    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
    98
    +0.20
    1.2%
    186
    +0.16
    0.9%
    358
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    317
    +0.20
    0.16
    407
    +0.16
    0.15
    358
    +0.12
    0.20
    Negative Logits
    ĻĤ
    -4.72
    Īĺ
    -3.78
    ±
    -3.37
    ĭ
    -3.36
    °
    -3.32
    Į
    -3.32
    Ļª
    -3.25
    ³
    -3.23
    ¤
    -3.22
    ľĵ
    -3.19
    POSITIVE LOGITS
     worldwide
    1.88
     globally
    1.68
     economically
    1.61
     technologies
    1.61
     economic
    1.59
     lucrative
    1.58
     infrastructure
    1.57
     nation
    1.56
     privat
    1.54
     widespread
    1.54
    Activations Density 5.278%

    No Known Activations