© 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. 13567
    Prev
    Next
    INDEX
    Explanations

    instances of the word "At."

    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
    280
    +0.17
    0.9%
    376
    +0.13
    0.7%
    243
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    457
    +0.17
    0.05
    243
    +0.13
    0.06
    33
    +0.11
    0.05
    Negative Logits
    ner
    -1.65
    η
    -1.57
    atrix
    -1.54
    iratory
    -1.50
    ners
    -1.47
    owitz
    -1.45
    er
    -1.42
    ands
    -1.41
     foods
    -1.39
    OWS
    -1.38
    POSITIVE LOGITS
    mosp
    3.06
     least
    2.64
    least
    2.49
    rophy
    2.10
    ĥ
    1.95
    kinson
    1.92
    Ļ
    1.81
     present
    1.81
    rial
    1.80
    omic
    1.79
    Activations Density 0.227%

    No Known Activations