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

    Neuronpedia

    APICircuit 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. 11712
    Prev
    Next
    INDEX
    Explanations

    mentions of different types of windows or window-related terms in the text

    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
    +0.23
    1.3%
    473
    +0.11
    0.7%
    211
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    211
    +0.23
    0.02
    148
    +0.11
    0.01
    473
    +0.11
    0.02
    Negative Logits
    ĵ
    -2.98
    ĥ½
    -2.95
    <|outofrange|>
    -2.70
                               
    -2.70
               
    -2.70
    ↵  ↵ 
    -2.70
     ↵     
    -2.70
     ↵           
    -2.70
                                                     
    -2.70
    ↵                                  
    -2.70
    POSITIVE LOGITS
     facing
    1.85
     viewed
    1.70
     faced
    1.61
    istor
    1.60
    $^
    1.59
     himself
    1.58
     sized
    1.56
    illa
    1.54
     height
    1.52
    ache
    1.50
    Activations Density 0.082%

    No Known Activations