© 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. 26795
    Prev
    Next
    INDEX
    Explanations

    subjects related to requests for information or questions in correspondence

    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
    369
    +0.17
    1.0%
    366
    +0.11
    0.6%
    159
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.17
    0.01
    202
    +0.11
    0.03
    159
    +0.10
    0.04
    Negative Logits
     operator
    -1.73
     defined
    -1.67
     pixels
    -1.57
     applied
    -1.57
    [@
    -1.48
     moments
    -1.47
     arrows
    -1.47
     juxt
    -1.44
     block
    -1.43
     operators
    -1.42
    POSITIVE LOGITS
    ĻĤ
    2.37
    debian
    1.59
    "));
    1.58
    enron
    1.57
    rangle
    1.56
    reement
    1.49
    Bye
    1.47
    "});
    1.46
    imbabwe
    1.46
    cuse
    1.46
    Activations Density 0.510%

    No Known Activations