© 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. 9140
    Prev
    Next
    INDEX
    Explanations

    conditional phrases or qualifiers in a legal or formal context

    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
    348
    +0.23
    1.4%
    303
    +0.13
    0.8%
    78
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.23
    -0.01
    78
    +0.13
    -0.01
    254
    +0.12
    0.01
    Negative Logits
     themselves
    -1.66
     herself
    -1.57
    ment
    -1.53
    ments
    -1.37
    ifies
    -1.28
    ('./
    -1.26
    resses
    -1.26
    istic
    -1.22
    :\
    -1.21
     itself
    -1.21
    POSITIVE LOGITS
    ¯
    4.28
    °
    4.07
    Ī
    4.00
    Ŀ
    3.97
    ¨
    3.96
    ª
    3.91
    Ļ
    3.91
    ī
    3.91
    ij
    3.87
    Ĥ¬
    3.86
    Activations Density 0.312%

    No Known Activations