© 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. 4-RES-SM
    11. 9164
    Prev
    Next
    INDEX
    Explanations

    various punctuation marks and formatting elements

    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
    21
    +0.35
    2.2%
    306
    +0.12
    0.7%
    316
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.35
    0.05
    316
    +0.12
    0.07
    103
    +0.12
    0.05
    Negative Logits
     affected
    -1.39
     aim
    -1.31
    forth
    -1.30
     recognise
    -1.30
     accord
    -1.29
    cot
    -1.28
    enden
    -1.27
     consecutive
    -1.27
     successive
    -1.26
     recognize
    -1.25
    POSITIVE LOGITS
    ĻĤ
    2.45
    ½
    1.98
    ı
    1.94
    ¿½
    1.94
    ģ
    1.93
    _>
    1.91
    ¼
    1.88
    »
    1.72
    ļ
    1.68
    Cause
    1.65
    Activations Density 0.599%

    No Known Activations