© 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. 28022
    Prev
    Next
    INDEX
    Explanations

    elements related to technology and media representation

    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
    261
    +0.14
    0.8%
    57
    +0.13
    0.8%
    271
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    261
    +0.14
    0.06
    57
    +0.13
    0.10
    497
    +0.12
    0.04
    Negative Logits
    taire
    -1.77
     exception
    -1.49
     regards
    -1.49
    apest
    -1.44
     admit
    -1.44
     trouble
    -1.41
     virtue
    -1.39
    liness
    -1.37
     Helsinki
    -1.37
    holds
    -1.37
    POSITIVE LOGITS
    Ļª
    3.48
    ĨĴ
    3.03
    ¦
    2.72
    į
    2.59
    º
    2.52
    ĭ
    2.49
    ³
    2.48
    ĺ
    2.48
    ¼
    2.46
    Ĵ
    2.45
    Activations Density 4.536%

    No Known Activations