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. MLP Post
    5. 0-MLP-SM
    6. 27410
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly closing parentheses

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    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
    23
    +0.30
    1.8%
    258
    +0.21
    1.2%
    250
    +0.20
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    250
    +0.30
    0.20
    412
    +0.21
    0.14
    258
    +0.20
    0.14
    Negative Logits
    ↵           
    -3.98
    ↵    ↵   
    -3.98
                                                                                   
    -3.98
                                                                                  
    -3.98
    ↵                                                                
    -3.98
    ↵↵↵       
    -3.98
                                         
    -3.98
                                                   
    -3.98
                                                                                                                                                                                                                                                                    
    -3.98
    ↵↵                   
    -3.98
    POSITIVE LOGITS
     marriages
    1.77
    wed
    1.75
     marrying
    1.71
     married
    1.68
     divorced
    1.61
    psy
    1.59
    between
    1.49
     marriage
    1.48
     intimate
    1.45
    ”.
    1.44
    Activations Density 5.441%

    No Known Activations