© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 32180
    Prev
    Next
    INDEX
    Explanations

    terms related to stabilization and management processes

    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
    47
    +0.22
    1.3%
    419
    +0.14
    0.8%
    72
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.22
    0.14
    400
    +0.14
    0.04
    419
    +0.13
    0.07
    Negative Logits
    ĸ´
    -3.42
    ±
    -3.37
    -3.34
    <|outofrange|>
    -3.34
    ↵↵
    -3.34
    ↵
    -3.34
    ↵
    -3.34
    -3.34
    -3.34
    č↵
    -3.34
    POSITIVE LOGITS
    ypse
    1.67
    msgstr
    1.58
    ously
    1.57
     itself
    1.56
     ourselves
    1.46
     starters
    1.43
    ptions
    1.43
    izations
    1.42
    ipes
    1.42
    apses
    1.42
    Activations Density 2.105%

    No Known Activations