© 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. 13074
    Prev
    Next
    INDEX
    Explanations

    expressions related to the concept of "sense" or perception, particularly in the context of emotions and awareness

    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
    314
    +0.14
    0.7%
    137
    +0.11
    0.6%
    48
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    198
    +0.14
    0.02
    314
    +0.11
    0.02
    127
    +0.11
    0.03
    Negative Logits
    cember
    -1.82
    cludes
    -1.62
    blockList
    -1.57
    itoring
    -1.52
    CLU
    -1.42
    ureus
    -1.41
    tsd
    -1.39
    ssä
    -1.39
    tawa
    -1.38
    uries
    -1.38
    POSITIVE LOGITS
    ably
    1.77
     amplifier
    1.68
    ingly
    1.64
    ier
    1.56
    asek
    1.55
    ional
    1.51
    isters
    1.48
    oud
    1.47
    orno
    1.44
    ilar
    1.42
    Activations Density 0.182%

    No Known Activations