© 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. 13433
    Prev
    Next
    INDEX
    Explanations

    negations and contrasts in mathematical or logical expressions

    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
    78
    +0.21
    1.3%
    397
    +0.16
    1.0%
    255
    +0.14
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.21
    0.00
    84
    +0.16
    0.01
    89
    +0.14
    0.01
    Negative Logits
    aries
    -1.78
    anges
    -1.63
     entire
    -1.57
    atically
    -1.51
    aneously
    -1.44
    atics
    -1.43
    ativity
    -1.41
     either
    -1.37
     correctly
    -1.36
     downloaded
    -1.36
    POSITIVE LOGITS
    ¬
    3.55
    Ĩ
    3.55
    ĥ½
    3.45
    ŀ
    3.45
    Ĥ¬
    3.42
    ī
    3.31
    ´
    3.29
    ¨
    3.27
    Īĺ
    3.17
    Ī
    3.16
    Activations Density 0.032%

    No Known Activations