© 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. 27638
    Prev
    Next
    INDEX
    Explanations

    auxillary verbs and expressions of possibility or questioning

    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
    478
    +0.12
    0.7%
    219
    +0.12
    0.7%
    153
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    219
    +0.12
    0.08
    103
    +0.12
    0.06
    2
    +0.11
    0.06
    Negative Logits
    Ļª
    -2.19
    liest
    -2.17
    ĨĴ
    -1.82
    ±
    -1.69
     against
    -1.59
    §
    -1.56
    ij
    -1.54
    oon
    -1.54
     stroke
    -1.53
    ump
    -1.52
    POSITIVE LOGITS
    LENGTH
    1.55
    golang
    1.52
    cerning
    1.49
    zens
    1.48
    lst
    1.46
    gnu
    1.41
    herty
    1.40
    pite
    1.37
    openssl
    1.37
    TeX
    1.37
    Activations Density 0.745%

    No Known Activations