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. 29206
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns and their associations

    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
    +0.13
    0.8%
    391
    +0.13
    0.8%
    51
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.13
    0.09
    51
    +0.13
    0.08
    191
    +0.12
    0.08
    Negative Logits
    aire
    -1.66
     their
    -1.62
    arium
    -1.61
     each
    -1.58
     him
    -1.54
     where
    -1.54
    break
    -1.50
     between
    -1.49
     these
    -1.48
    wise
    -1.47
    POSITIVE LOGITS
     own
    3.09
     Majesty
    2.32
    panic
    2.13
     wife
    2.11
    sing
    1.99
     Own
    1.93
     Excell
    1.92
     hometown
    1.88
    elf
    1.84
     girlfriend
    1.81
    Activations Density 0.414%

    No Known Activations