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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 29073
    Prev
    Next
    INDEX
    Explanations

    references to licenses or license-related terms

    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
    205
    +0.14
    0.8%
    497
    +0.12
    0.7%
    231
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    329
    +0.14
    0.01
    205
    +0.12
    0.01
    315
    +0.11
    0.01
    Negative Logits
    Īĺ
    -4.38
            ↵ 
    -4.23
    ↵                       
    -4.23
                                                                    
    -4.23
    ↵            
    -4.23
                                                  
    -4.23
                         
    -4.23
                                                                               
    -4.23
    ↵  Âł
    -4.23
    ↵                                                                                                                                
    -4.23
    POSITIVE LOGITS
     Appeal
    1.31
    ôt
    1.30
    arser
    1.28
     dire
    1.27
     Agreement
    1.25
    seed
    1.21
     agreement
    1.19
    ieur
    1.19
    IEW
    1.18
    ÑĮ
    1.18
    Activations Density 0.060%

    No Known Activations