© 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. 4-RES-SM
    11. 9398
    Prev
    Next
    INDEX
    Explanations

    references to geopolitical entities and military actions related to international relations

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    307
    +0.24
    1.4%
    306
    +0.17
    1.0%
    156
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    168
    +0.24
    0.12
    98
    +0.17
    0.06
    418
    +0.14
    0.07
    Negative Logits
    ktop
    -1.64
    ))^
    -1.61
    "}](#
    -1.52
    arynge
    -1.51
    sible
    -1.50
    lace
    -1.47
    ,...,
    -1.47
    naments
    -1.47
     etc
    -1.45
    ãģ¾ãģĽ
    -1.44
    POSITIVE LOGITS
    Ī
    3.27
    ģ
    3.22
    ħ
    3.13
    ¿½
    3.08
    Ĥ
    3.08
    Ĩ
    3.07
    ¿
    3.03
    ĵ
    2.98
    ²
    2.96
    2.96
    Activations Density 1.489%

    No Known Activations