© 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. 13516
    Prev
    Next
    INDEX
    Explanations

    phrases related to political power and influential figures

    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
    399
    +0.12
    0.7%
    469
    +0.10
    0.5%
    330
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    399
    +0.12
    0.15
    345
    +0.10
    0.21
    421
    +0.09
    0.12
    Negative Logits
    [])
    -1.42
    Expected
    -1.40
     worth
    -1.33
    ='
    -1.31
    ialize
    -1.29
    ~~~
    -1.25
    )'
    -1.23
    void
    -1.22
    thon
    -1.22
    shoot
    -1.20
    POSITIVE LOGITS
    ·¸
    2.53
    ļ
    2.47
    ĻĤ
    2.28
    ³
    2.26
    ŀ
    2.26
    Ĥ¬
    2.25
    ±
    2.25
    ¿½
    2.22
    ł
    2.19
    ↵
    2.15
    Activations Density 10.050%

    No Known Activations