© 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. 1-RES-SM
    11. 24362
    Prev
    Next
    INDEX
    Explanations

    references to specific places or locations associated with political or social events

    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
    171
    +0.15
    0.8%
    195
    +0.14
    0.8%
    200
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    171
    +0.15
    0.07
    410
    +0.14
    0.06
    348
    +0.13
    0.06
    Negative Logits
    ([]
    -1.70
    fulness
    -1.51
    of
    -1.47
    oretical
    -1.40
     flooding
    -1.39
    acting
    -1.39
    alling
    -1.39
    standing
    -1.37
    ights
    -1.35
    saving
    -1.33
    POSITIVE LOGITS
    ĻĤ
    2.56
    ķ
    2.28
    ģ
    2.27
    Ķ
    2.16
    Ĩ
    2.11
    ½
    2.07
    Į
    2.06
    ¹
    2.03
    ĸ
    2.00
    ¡
    1.95
    Activations Density 0.796%

    No Known Activations