© 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. 27954
    Prev
    Next
    INDEX
    Explanations

    legal references and citations in text

    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
    294
    +0.14
    0.8%
    456
    +0.11
    0.6%
    266
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    294
    +0.14
    0.04
    374
    +0.11
    0.03
    254
    +0.10
    0.04
    Negative Logits
    riber
    -1.68
    áĢ¬
    -1.51
    $$
    -1.46
     quo
    -1.39
    à±ģ
    -1.38
    vez
    -1.35
    asma
    -1.35
    /@
    -1.33
     donors
    -1.31
    \$
    -1.30
    POSITIVE LOGITS
    ĻĤ
    3.72
    Īĺ
    3.46
    ¦
    3.10
    ¯
    3.00
    ²
    2.87
    «
    2.79
    ´
    2.79
    ±
    2.65
    Ĥ¬
    2.62
    IJ
    2.60
    Activations Density 0.097%

    No Known Activations