© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 32164
    Prev
    Next
    INDEX
    Explanations

    specific terms related to materials and their properties in various 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
    47
    +0.14
    0.8%
    217
    +0.14
    0.8%
    302
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    66
    +0.14
    0.09
    419
    +0.14
    0.07
    101
    +0.12
    0.04
    Negative Logits
    amais
    -1.46
    Ŀ
    -1.45
    ilon
    -1.45
    anas
    -1.38
    STRA
    -1.32
     Case
    -1.30
    ·
    -1.29
     pla
    -1.26
     Page
    -1.26
     Synd
    -1.25
    POSITIVE LOGITS
     nature
    1.75
    nger
    1.68
    IPE
    1.61
    ctomy
    1.60
     arises
    1.60
    andum
    1.55
    gue
    1.52
    vist
    1.45
     itself
    1.45
    chol
    1.43
    Activations Density 1.105%

    No Known Activations