© 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. 14102
    Prev
    Next
    INDEX
    Explanations

    content related to education, educational programs, and learning opportunities

    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
    451
    +0.13
    0.8%
    345
    +0.13
    0.7%
    135
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    418
    +0.13
    0.03
    186
    +0.13
    0.04
    72
    +0.13
    0.03
    Negative Logits
    ij
    -3.21
    ĥ½
    -2.85
    Ģ
    -2.84
    ·¸
    -2.74
    ³
    -2.73
    Ĵ
    -2.72
    Ĥ¬
    -2.66
    Ļª
    -2.65
    º
    -2.64
    ĻĤ
    -2.61
    POSITIVE LOGITS
     attainment
    2.33
     programs
    2.25
     outreach
    2.24
     institutions
    2.23
     teacher
    2.19
     curriculum
    2.12
     programmes
    2.11
     curric
    2.11
     opportunities
    2.07
    ally
    2.01
    Activations Density 0.347%

    No Known Activations