© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 2-RES-SM
    11. 28397
    Prev
    Next
    INDEX
    Explanations

    phrases indicating geographical locations

    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
    247
    +0.11
    0.6%
    316
    +0.10
    0.6%
    115
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    413
    +0.11
    0.01
    228
    +0.10
    0.02
    236
    +0.10
    0.02
    Negative Logits
    Īĺ
    -3.35
    ¢
    -3.20
    ¥
    -3.18
    ĥ½
    -3.15
    Ł
    -3.11
    Ĥ
    -3.07
    ŀ
    -3.06
    «
    -2.96
    ¹
    -2.95
    Ŀ
    -2.94
    POSITIVE LOGITS
     adjacent
    2.01
     nearby
    1.97
     nowhere
    1.97
     upstream
    1.96
     near
    1.86
     west
    1.85
     south
    1.82
     east
    1.81
     north
    1.77
     atop
    1.75
    Activations Density 0.117%

    No Known Activations