© 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. 4-RES-SM
    6. 11931
    Prev
    Next
    INDEX
    Explanations

    features and specifications of products, particularly in descriptions related to home furnishings and real estate

    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
    25
    +0.16
    0.9%
    269
    +0.12
    0.7%
    198
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    462
    +0.16
    0.11
    401
    +0.12
    0.06
    179
    +0.11
    0.12
    Negative Logits
    ↵                 
    -8.94
    ↵             
    -8.94
                                                      
    -8.94
    ↵  Âł
    -8.94
    ↵↵      
    -8.94
    <|outofrange|>
    -8.94
             
    -8.94
    ↵                                         
    -8.94
    ↵                                
    -8.94
    ↵        ↵       
    -8.94
    POSITIVE LOGITS
     ensures
    2.57
     blend
    2.04
     added
    2.00
     ensured
    2.00
     plus
    2.00
     ensure
    1.91
     lets
    1.90
     featuring
    1.89
     adds
    1.89
     allows
    1.87
    Activations Density 2.523%

    No Known Activations