© 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. 9270
    Prev
    Next
    INDEX
    Explanations

    references to Microsoft and related schemas

    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
    118
    +0.14
    0.8%
    478
    +0.12
    0.7%
    129
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    118
    +0.14
    0.02
    349
    +0.12
    0.02
    255
    +0.10
    -0.00
    Negative Logits
     little
    -1.52
    iginally
    -1.50
    inates
    -1.41
    olate
    -1.38
    LINE
    -1.38
    few
    -1.37
     locality
    -1.33
    icate
    -1.30
     cease
    -1.29
     compose
    -1.28
    POSITIVE LOGITS
    ī
    2.13
    Ģ
    2.10
    ĻĤ
    2.09
    ¬
    1.90
    ļ
    1.90
    ®
    1.85
    ©
    1.82
    ↵↵
    1.79
                ↵
    1.79
    1.79
    Activations Density 0.107%

    No Known Activations