© 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. 13096
    Prev
    Next
    INDEX
    Explanations

    specific terms or phrases related to cultural identities or affiliations

    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
    153
    +0.17
    0.9%
    5
    +0.15
    0.8%
    98
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.17
    0.02
    271
    +0.15
    -0.01
    210
    +0.13
    0.04
    Negative Logits
    ably
    -1.45
    ously
    -1.41
     formalin
    -1.38
    ified
    -1.36
    entional
    -1.30
     “
    -1.29
    onuclear
    -1.28
    iosity
    -1.28
    curs
    -1.27
    isance
    -1.27
    POSITIVE LOGITS
     Pradesh
    1.89
     Suite
    1.78
    mania
    1.57
    öl
    1.57
    stown
    1.49
    essee
    1.48
    land
    1.46
    alian
    1.45
     Wales
    1.44
     River
    1.41
    Activations Density 2.152%

    No Known Activations