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. 11882
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "take" and its various forms

    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
    32
    +0.11
    0.6%
    396
    +0.11
    0.6%
    472
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    32
    +0.11
    0.04
    472
    +0.11
    0.06
    89
    +0.11
    0.04
    Negative Logits
    edly
    -2.12
    ĨĴ
    -2.07
    ¶
    -1.82
    ¬
    -1.78
    ľ
    -1.76
    »
    -1.75
    Ń
    -1.73
    astically
    -1.73
    §
    -1.69
    hood
    -1.63
    POSITIVE LOGITS
     advantage
    2.47
    aways
    2.25
     care
    2.02
    uchi
    1.80
    ership
    1.79
    away
    1.78
     heed
    1.76
     note
    1.75
     cue
    1.71
     turns
    1.70
    Activations Density 0.218%

    No Known Activations