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. 29452
    Prev
    Next
    INDEX
    Explanations

    references to animals and their descriptions

    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
    293
    +0.17
    0.9%
    507
    +0.14
    0.8%
    47
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    81
    +0.17
    0.12
    100
    +0.14
    0.15
    256
    +0.13
    0.16
    Negative Logits
    Ĥ¬
    -1.93
    ¾
    -1.84
    ĸ´
    -1.83
    ľĵ
    -1.79
    ľ
    -1.79
    ı
    -1.71
    ŀ
    -1.66
    Ŀ
    -1.62
    ij
    -1.60
    ¯
    -1.59
    POSITIVE LOGITS
     trillion
    1.46
     driven
    1.42
    ivism
    1.31
    cible
    1.31
     composed
    1.28
    etic
    1.27
     emig
    1.26
    ivist
    1.25
     living
    1.25
     travelled
    1.24
    Activations Density 5.958%

    No Known Activations