© 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. 12466
    Prev
    Next
    INDEX
    Explanations

    references to specific locations or places

    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
    369
    +0.16
    0.9%
    314
    +0.13
    0.7%
    272
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.16
    0.01
    314
    +0.13
    0.02
    181
    +0.12
    0.02
    Negative Logits
    rias
    -1.90
    cess
    -1.82
    thon
    -1.75
    boarding
    -1.73
    atra
    -1.71
    ria
    -1.70
    arma
    -1.70
    board
    -1.66
    lla
    -1.64
    ching
    -1.58
    POSITIVE LOGITS
    aze
    1.70
    asgow
    1.63
    ings
    1.51
    ued
    1.47
    ucose
    1.47
    inger
    1.39
    stones
    1.30
    ossier
    1.29
    ian
    1.29
     reflect
    1.27
    Activations Density 0.105%

    No Known Activations