© 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 EvalsExportsAPI 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. 12694
    Prev
    Next
    INDEX
    Explanations

    terms related to engagement and participation

    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
    +0.14
    0.8%
    206
    +0.14
    0.8%
    255
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.14
    0.03
    386
    +0.14
    0.01
    443
    +0.13
    0.02
    Negative Logits
    pora
    -1.88
    thood
    -1.55
    chers
    -1.45
    quin
    -1.42
    ystems
    -1.42
    cribing
    -1.39
     code
    -1.35
     utmost
    -1.34
    iginally
    -1.33
    phy
    -1.32
    POSITIVE LOGITS
    Ģ
    3.06
    ĥ½
    2.82
    ª
    2.67
    lish
    2.58
    ³
    2.52
    Ĺ
    2.51
    ¡
    2.47
    ±
    2.44
    Ī
    2.42
    agement
    2.32
    Activations Density 0.218%

    No Known Activations