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. 12188
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "to"

    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
    478
    +0.22
    1.2%
    69
    +0.17
    0.9%
    142
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.22
    0.14
    434
    +0.17
    0.09
    142
    +0.12
    0.09
    Negative Logits
    arity
    -2.08
    er
    -1.90
    hole
    -1.88
    arily
    -1.88
    eous
    -1.88
    liness
    -1.76
    ly
    -1.62
    icular
    -1.61
    ishly
    -1.61
    teen
    -1.60
    POSITIVE LOGITS
    pless
    2.21
    ĥ
    2.18
    asts
    2.14
    ľĵ
    2.07
    ĵ
    1.87
    ±
    1.77
    ↵                                          
    1.77
                                          
    1.77
    ↵  âĢĥ
    1.77
    ↵                  
    1.77
    Activations Density 0.891%

    No Known Activations