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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 28811
    Prev
    Next
    INDEX
    Explanations

    references to Christmas and holiday-related themes

    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
    +0.24
    1.4%
    56
    +0.15
    0.9%
    93
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.24
    0.00
    145
    +0.15
    0.05
    38
    +0.11
    0.03
    Negative Logits
    Īĺ
    -4.67
    ı
    -4.41
    ij
    -4.28
     ↵         
    -4.27
    ↵                                                                                                
    -4.27
    <|outofrange|>
    -4.27
    ↵↵            
    -4.27
    ↵  
    -4.27
                ↵ 
    -4.27
    ↵	     
    -4.27
    POSITIVE LOGITS
     celebrations
    2.44
     celebration
    2.19
     season
    2.14
     decorations
    2.08
     ceremonies
    2.04
     Lights
    2.01
     bells
    1.97
     holiday
    1.96
    flower
    1.93
    iversary
    1.92
    Activations Density 0.465%

    No Known Activations