© 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. 29267
    Prev
    Next
    INDEX
    Explanations

    terms and concepts related to legal and contractual matters

    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
    23
    +0.20
    1.1%
    156
    +0.13
    0.7%
    472
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.20
    0.12
    365
    +0.13
    0.08
    472
    +0.13
    0.09
    Negative Logits
    ®
    -4.44
    -4.38
    ↵↵
    -4.38
    ↵↵
    -4.38
    <|outofrange|>
    -4.38
    ↵
    -4.38
    -4.38
    -4.38
    ↵
    -4.38
                  ↵
    -4.38
    POSITIVE LOGITS
    ographers
    1.60
     itself
    1.57
    '?"
    1.53
    owners
    1.49
    leading
    1.47
    then
    1.44
     blacks
    1.43
    going
    1.42
     themselves
    1.41
     difference
    1.34
    Activations Density 1.580%

    No Known Activations