© 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. MLP Post
    5. 2-MLP-SM
    6. 25781
    Prev
    Next
    INDEX
    Explanations

    specific phrases or constructs related to genetic conditions and their implications

    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/2-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.2.hook_mlp_out
    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
    453
    +0.14
    0.8%
    217
    +0.13
    0.7%
    359
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    217
    +0.14
    0.11
    334
    +0.13
    0.06
    293
    +0.12
    0.05
    Negative Logits
    ¶
    -2.67
    ĨĴ
    -2.66
    Ĭ
    -2.63
    ·
    -2.54
           
    -2.53
     ↵      
    -2.53
                  
    -2.53
    čč         
    -2.53
                         
    -2.53
                                                                                                                                                                                                                                    
    -2.53
    POSITIVE LOGITS
    pntd
    1.67
     yourselves
    1.50
     knowledgeable
    1.49
    ERT
    1.48
     qualified
    1.48
     timely
    1.47
    Docket
    1.47
    aintiff
    1.46
    actively
    1.42
    IOR
    1.41
    Activations Density 2.214%

    No Known Activations