© 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. 14228
    Prev
    Next
    INDEX
    Explanations

    information related to disclaimers and usage terms

    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
    348
    +0.16
    0.9%
    78
    +0.14
    0.8%
    306
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    289
    +0.16
    0.02
    150
    +0.14
    0.04
    334
    +0.13
    0.04
    Negative Logits
    lotte
    -1.91
    istics
    -1.72
    rapeutics
    -1.65
    phalt
    -1.58
    pired
    -1.54
    plicity
    -1.45
    istical
    -1.41
    inki
    -1.36
     occupancy
    -1.35
     stride
    -1.31
    POSITIVE LOGITS
     pursuant
    2.25
     by
    1.87
     under
    1.83
     WITHOUT
    1.81
     knowingly
    1.77
    ){#
    1.76
     herein
    1.76
     unto
    1.74
     solely
    1.71
     against
    1.63
    Activations Density 0.271%

    No Known Activations