© 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. 31396
    Prev
    Next
    INDEX
    Explanations

    mentions of rural and urban contexts, particularly focusing on the distinction and issues related to rural areas

    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
    156
    +0.25
    1.5%
    505
    +0.10
    0.6%
    59
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    276
    +0.25
    0.02
    145
    +0.10
    0.03
    505
    +0.10
    0.02
    Negative Logits
    ção
    -1.62
    ções
    -1.52
     Ambassador
    -1.45
    lessly
    -1.41
     subpo
    -1.40
    ÑĢоÑģ
    -1.40
    radiol
    -1.40
     vacc
    -1.38
     true
    -1.37
     suit
    -1.36
    POSITIVE LOGITS
    ĸ
    2.14
    ¢
    2.09
    ĵ
    2.08
    ¡
    2.07
    Ł
    2.00
    iens
    1.94
    Ĺ
    1.91
    µ
    1.90
    idal
    1.90
    Ĵ
    1.87
    Activations Density 0.183%

    No Known Activations