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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 28473
    Prev
    Next
    INDEX
    Explanations

    instances of the word "having" and its variations in various 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
    27
    +0.12
    0.7%
    123
    +0.11
    0.6%
    436
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    108
    +0.12
    0.06
    358
    +0.11
    0.06
    27
    +0.10
    0.05
    Negative Logits
    Ļª
    -2.98
    ľ
    -2.88
    ij
    -2.81
    §
    -2.79
    Ń
    -2.70
    ŀ
    -2.70
    ĸ
    -2.66
    ¥
    -2.62
    ļ
    -2.62
    Ĺ
    -2.62
    POSITIVE LOGITS
     dinner
    1.79
     lately
    1.66
     intercourse
    1.61
     fun
    1.55
     forgotten
    1.54
    cele
    1.54
    ctic
    1.52
    ymoon
    1.52
     stolen
    1.49
     conversations
    1.47
    Activations Density 0.299%

    No Known Activations