Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 29140
    Prev
    Next
    INDEX
    Explanations

    references to specific types of food or dietary practices, particularly related to ketogenic diets

    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
    +0.20
    1.1%
    369
    +0.19
    1.1%
    299
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    299
    +0.20
    0.04
    369
    +0.19
    0.02
    415
    +0.13
    0.02
    Negative Logits
    TRODUCTION
    -1.71
    herence
    -1.63
    axel
    -1.58
    cess
    -1.57
    geries
    -1.53
    cession
    -1.52
    herent
    -1.52
    googleapis
    -1.47
    ventions
    -1.43
    fact
    -1.43
    POSITIVE LOGITS
    Ĵ
    2.22
    ī
    2.11
    afka
    1.75
    ļ
    1.72
    Ģ
    1.71
    ovo
    1.69
    ĨĴ
    1.65
    ansas
    1.63
    dar
    1.62
    irk
    1.62
    Activations Density 0.081%

    No Known Activations