© 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. 30760
    Prev
    Next
    INDEX
    Explanations

    technical terms and constructs in programming and data organization

    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
    326
    +0.22
    1.2%
    237
    +0.13
    0.8%
    499
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.22
    0.07
    237
    +0.13
    0.13
    181
    +0.13
    0.10
    Negative Logits
     tomorrow
    -1.77
     evening
    -1.48
     side
    -1.43
     overnight
    -1.38
    bard
    -1.35
     afternoon
    -1.34
     rounds
    -1.32
    ttes
    -1.31
     minutes
    -1.31
     reservoirs
    -1.30
    POSITIVE LOGITS
    ait
    1.39
     territory
    1.39
    ition
    1.38
    proved
    1.38
    UIC
    1.32
    apon
    1.26
    liked
    1.26
    ativistic
    1.25
     treason
    1.24
    cher
    1.21
    Activations Density 4.634%

    No Known Activations