© 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. 31108
    Prev
    Next
    INDEX
    Explanations

    references to legal cases and their details

    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
    419
    +0.19
    1.1%
    201
    +0.13
    0.7%
    143
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.19
    0.21
    156
    +0.13
    0.24
    71
    +0.12
    0.24
    Negative Logits
    .[]{
    -1.62
    `.
    -1.58
    ).](
    -1.57
    $.[]{
    -1.52
    )',
    -1.45
    ######
    -1.43
     benchmark
    -1.43
    ]{};
    -1.42
    )."
    -1.38
    ."""
    -1.38
    POSITIVE LOGITS
     ribs
    1.37
     attachments
    1.35
     contact
    1.34
    olateral
    1.32
    acer
    1.31
     Petitioner
    1.30
    ¿½
    1.28
    ĻĤ
    1.26
     paras
    1.25
     Hait
    1.22
    Activations Density 5.271%

    No Known Activations