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. Residuals
    5. 2-RES-SM
    6. 29208
    Prev
    Next
    INDEX
    Explanations

    references to legal terminology and regulations

    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
    369
    +0.15
    0.8%
    156
    +0.14
    0.7%
    233
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    429
    +0.15
    0.02
    215
    +0.14
    0.03
    360
    +0.12
    0.02
    Negative Logits
    ·¸
    -2.19
    TRODUCTION
    -1.76
    ção
    -1.75
    izumab
    -1.72
    ffect
    -1.72
    cology
    -1.66
    ↵                     
    -1.64
    ↵                                           
    -1.64
    <|outofrange|>
    -1.64
    <|outofrange|>
    -1.64
    POSITIVE LOGITS
    derr
    1.84
     up
    1.74
    ups
    1.63
    ings
    1.61
    oke
    1.60
    oked
    1.59
    encil
    1.54
    uart
    1.52
    ellate
    1.51
    ede
    1.46
    Activations Density 0.056%

    No Known Activations