© 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. 4-RES-SM
    11. 10503
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "today" and its variations, often in the context of timing or relevance

    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
    29
    +0.14
    0.8%
    56
    +0.13
    0.7%
    498
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    269
    +0.14
    0.04
    56
    +0.13
    0.02
    33
    +0.12
    0.03
    Negative Logits
    ĻĤ
    -4.42
    Ļª
    -4.25
    ŀ
    -4.18
    ·¸
    -4.12
    Ĩ
    -4.05
                ↵
    -3.88
    -3.88
    ↵
    -3.88
    -3.88
    <|outofrange|>
    -3.88
    POSITIVE LOGITS
    's
    2.74
     afternoon
    2.25
    ’
    2.06
    ÃŃs
    2.04
     morning
    2.01
     evening
    1.89
    after
    1.77
    ÂĴ
    1.73
    abouts
    1.71
     announced
    1.66
    Activations Density 0.249%

    No Known Activations