© 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. 14337
    Prev
    Next
    INDEX
    Explanations

    details and specifications related to legal and procedural contexts

    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
    306
    +0.18
    1.3%
    307
    +0.15
    1.1%
    337
    +0.14
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    306
    +0.18
    0.05
    174
    +0.15
    0.06
    337
    +0.14
    0.07
    Negative Logits
    ¼
    -1.55
    ķ
    -1.28
    ¥
    -1.28
    Ŀ
    -1.27
    IJ
    -1.23
    »¿
    -1.20
    Ł
    -1.19
    ary
    -1.19
    Ļ
    -1.16
    naments
    -1.16
    POSITIVE LOGITS
    à¬
    1.43
    à½
    1.37
    à´
    1.36
    ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
    1.31
    à²
    1.27
    àº
    1.27
    ãĥ³ãĥĪ
    1.25
    áŀ
    1.25
    à¨
    1.23
     economy
    1.22
    Activations Density 2.298%

    No Known Activations