© 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. 27448
    Prev
    Next
    INDEX
    Explanations

    words related to the color pink or things described with "ink"

    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
    347
    +0.15
    0.8%
    108
    +0.12
    0.7%
    10
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    347
    +0.15
    0.03
    66
    +0.12
    0.03
    260
    +0.11
    0.02
    Negative Logits
    Īĺ
    -2.35
    ²
    -2.10
    Ľ
    -2.08
    ĵ
    -2.06
    ĥ½
    -2.05
    ĺ
    -2.02
    Ń
    -2.00
    Ł
    -1.91
    ³
    -1.89
    ·
    -1.86
    POSITIVE LOGITS
    elijk
    2.03
    wort
    1.97
    ernel
    1.97
    age
    1.88
    manship
    1.84
    ish
    1.77
    horn
    1.76
    lace
    1.75
    heart
    1.71
    back
    1.61
    Activations Density 0.184%

    No Known Activations