© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 26475
    Prev
    Next
    INDEX
    Explanations

    phrases related to online platforms or applications

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    501
    +0.12
    0.7%
    420
    +0.12
    0.7%
    441
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    420
    +0.12
    0.01
    120
    +0.12
    0.01
    394
    +0.11
    0.02
    Negative Logits
    rbrack
    -1.64
     nervous
    -1.50
     victim
    -1.45
     allergic
    -1.44
     civilian
    -1.44
     slightest
    -1.42
    patient
    -1.39
     newly
    -1.38
     living
    -1.38
     outward
    -1.38
    POSITIVE LOGITS
    º
    2.73
    ĻĤ
    2.63
    ģ
    2.57
    ŀ
    2.48
    ī
    2.46
    İ
    2.42
    Ł
    2.32
    ®
    2.26
    Ĭ
    2.23
    ·
    2.23
    Activations Density 0.084%

    No Known Activations