© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 11292
    Prev
    Next
    INDEX
    Explanations

    concepts related to health, nutrition, and the impacts of lifestyle choices on well-being

    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
    156
    +0.24
    1.4%
    406
    +0.13
    0.7%
    93
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    93
    +0.24
    0.14
    392
    +0.13
    0.13
    287
    +0.10
    0.15
    Negative Logits
    ↵
    -5.17
    ↵↵
    -5.17
    <|outofrange|>
    -5.17
    ↵
    -5.17
    -5.17
    <|outofrange|>
    -5.17
    č↵
    -5.17
    <|outofrange|>
    -5.17
    č↵
    -5.17
    ↵    ↵
    -5.17
    POSITIVE LOGITS
    "}](#
    2.02
    haul
    1.59
    fall
    1.56
    pass
    1.46
    Nil
    1.46
    ician
    1.43
    necessary
    1.42
    pile
    1.41
    fecture
    1.40
     defeating
    1.40
    Activations Density 3.181%

    No Known Activations