© 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. 12769
    Prev
    Next
    INDEX
    Explanations

    references to crystal-related objects or materials

    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
    371
    +0.12
    0.7%
    438
    +0.11
    0.6%
    79
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    177
    +0.12
    0.01
    448
    +0.11
    0.01
    173
    +0.11
    0.01
    Negative Logits
     person
    -1.79
     succeeding
    -1.77
    akin
    -1.67
     miss
    -1.57
    quit
    -1.54
     happ
    -1.51
     vain
    -1.49
     unsuccessful
    -1.48
    cess
    -1.45
    eligible
    -1.44
    POSITIVE LOGITS
    istry
    1.90
    ización
    1.74
    isión
    1.67
     symmetry
    1.60
    ↵³³
    1.59
    edu
    1.55
    vre
    1.55
     representation
    1.54
     symmetries
    1.54
     discloses
    1.53
    Activations Density 0.129%

    No Known Activations