© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 11517
    Prev
    Next
    INDEX
    Explanations

    legal terminology and references related to official documents and cases

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    291
    +0.14
    0.8%
    109
    +0.12
    0.7%
    216
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    169
    +0.14
    0.09
    400
    +0.12
    0.05
    253
    +0.12
    0.09
    Negative Logits
    ↵
    -3.35
               ↵
    -3.35
    -3.35
    ↵
    -3.35
    ↵  ↵
    -3.35
    <|outofrange|>
    -3.35
    -3.35
    ↵  Âł
    -3.35
    ↵
    -3.35
    -3.35
    POSITIVE LOGITS
     OF
    2.96
     COMPANY
    2.82
    ITY
    2.81
     CIRCUIT
    2.77
     DISTRICT
    2.60
     AND
    2.56
    FIELD
    2.55
    ATION
    2.54
     SOFTWARE
    2.49
     FOR
    2.49
    Activations Density 0.553%

    No Known Activations