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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 30742
    Prev
    Next
    INDEX
    Explanations

    mentions of substrates in scientific 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
    +0.22
    1.3%
    169
    +0.12
    0.7%
    389
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    343
    +0.22
    0.01
    389
    +0.12
    0.01
    17
    +0.11
    0.00
    Negative Logits
    ĥ½
    -3.93
    ŀ
    -3.10
    ·
    -3.06
    ĻĤ
    -2.97
    Ń
    -2.96
    ³
    -2.94
    ±
    -2.84
    §
    -2.81
    ¾
    -2.76
    ¸
    -2.72
    POSITIVE LOGITS
    ization
    1.69
    izable
    1.68
    manship
    1.67
    pile
    1.62
    iors
    1.56
    ivity
    1.54
     sensed
    1.53
    tern
    1.52
     coated
    1.51
     needs
    1.48
    Activations Density 0.065%

    No Known Activations