© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 11406
    Prev
    Next
    INDEX
    Explanations

    references to user interface elements in programming contexts

    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
    156
    +0.17
    0.9%
    358
    +0.14
    0.8%
    409
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    173
    +0.17
    0.02
    443
    +0.14
    0.02
    411
    +0.10
    0.02
    Negative Logits
    ilities
    -1.59
     dehydrogenase
    -1.57
    xygen
    -1.57
    ylase
    -1.57
     drinking
    -1.52
     bacteria
    -1.52
    ught
    -1.51
     dormant
    -1.49
     little
    -1.49
     neither
    -1.48
    POSITIVE LOGITS
     Caption
    2.29
    port
    2.09
    iew
    1.78
     seja
    1.67
    ista
    1.66
    Holder
    1.66
    ership
    1.64
    er
    1.63
    rera
    1.57
    acional
    1.55
    Activations Density 0.084%

    No Known Activations