© 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. 12920
    Prev
    Next
    INDEX
    Explanations

    references to legal, formal, or ceremonial contexts

    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
    54
    +0.14
    0.8%
    47
    +0.13
    0.7%
    401
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    401
    +0.14
    0.11
    123
    +0.13
    0.13
    130
    +0.12
    0.14
    Negative Logits
     enough
    -1.53
     proofs
    -1.48
    mente
    -1.47
     him
    -1.43
     Thunder
    -1.42
     regards
    -1.41
     Access
    -1.40
     Essays
    -1.39
     Students
    -1.38
     Partners
    -1.36
    POSITIVE LOGITS
    hores
    1.60
    ()</
    1.56
    ciparum
    1.54
    $–
    1.53
    orbent
    1.51
    idase
    1.50
    enter
    1.48
    iratory
    1.48
    ogram
    1.46
    cies
    1.44
    Activations Density 3.433%

    No Known Activations