© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 10675
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate recommendations or suggestions

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    307
    +0.16
    0.9%
    354
    +0.15
    0.8%
    453
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    485
    +0.16
    0.05
    354
    +0.15
    0.05
    330
    +0.11
    0.06
    Negative Logits
    į
    -2.37
    ĻĤ
    -2.22
    Ģ
    -2.06
    »¿
    -2.05
    «
    -1.96
    Ľ
    -1.79
    ĺ
    -1.75
    ±
    -1.71
    Ħ
    -1.69
    İ
    -1.66
    POSITIVE LOGITS
     adding
    2.43
     contacting
    2.24
     upgrading
    2.23
     joining
    2.22
     removing
    2.20
     alternatives
    2.15
     replacing
    2.12
     installing
    2.10
     applying
    2.10
     ways
    2.09
    Activations Density 0.565%

    No Known Activations