© 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. 1-RES-SM
    11. 24236
    Prev
    Next
    INDEX
    Explanations

    special characters and diacritics in words

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    410
    +0.23
    1.4%
    111
    +0.16
    1.0%
    203
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    428
    +0.23
    0.04
    5
    +0.16
    0.04
    205
    +0.12
    0.04
    Negative Logits
     prone
    -1.68
    olation
    -1.58
     shown
    -1.52
    aten
    -1.50
    ");
    -1.50
     causing
    -1.49
      ↵
    -1.48
     middle
    -1.46
    Ĺ
    -1.44
    ired
    -1.44
    POSITIVE LOGITS
    ties
    1.76
    tre
    1.68
    cif
    1.64
    stad
    1.59
    že
    1.56
    minist
    1.54
    cias
    1.52
    ynote
    1.49
    ncia
    1.47
    liche
    1.46
    Activations Density 0.944%

    No Known Activations