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. 29413
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to energy consumption and efficiency

    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
    269
    +0.17
    1.0%
    156
    +0.15
    0.9%
    219
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    269
    +0.17
    0.03
    38
    +0.15
    0.01
    219
    +0.13
    0.02
    Negative Logits
     adoles
    -1.82
     tears
    -1.77
    ieurs
    -1.74
    ieur
    -1.69
     apologize
    -1.68
     mistrial
    -1.68
    aughters
    -1.66
     hers
    -1.65
    iquit
    -1.63
     rude
    -1.61
    POSITIVE LOGITS
    scape
    1.80
    seek
    1.74
     reserves
    1.74
    store
    1.71
    zone
    1.69
    core
    1.67
    corp
    1.65
     emitted
    1.64
    éĩı
    1.63
     consumption
    1.63
    Activations Density 0.097%

    No Known Activations