© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 4-RES-SM
    6. 14588
    Prev
    Next
    INDEX
    Explanations

    terms related to unsustainable practices

    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
    192
    +0.13
    0.7%
    21
    +0.12
    0.7%
    49
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    192
    +0.13
    0.01
    386
    +0.12
    0.01
    31
    +0.11
    0.03
    Negative Logits
    rile
    -1.89
    zes
    -1.54
    ment
    -1.51
    shire
    -1.49
    tion
    -1.48
    ments
    -1.44
    doms
    -1.44
    ria
    -1.42
     products
    -1.41
     dinner
    -1.40
    POSITIVE LOGITS
    creen
    2.31
    leep
    2.25
    ynchron
    2.25
    ¢
    2.24
    «
    2.23
    chool
    2.20
    ymmetric
    2.19
    Ĥ¬
    2.17
    ·
    2.17
    µ
    2.16
    Activations Density 0.179%

    No Known Activations