© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 26264
    Prev
    Next
    INDEX
    Explanations

    references to loss and grief

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    474
    +0.12
    0.7%
    324
    +0.10
    0.6%
    185
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    474
    +0.12
    0.02
    239
    +0.10
    0.03
    215
    +0.10
    0.03
    Negative Logits
    Ñĥм
    -1.51
    acon
    -1.48
    utor
    -1.46
    cco
    -1.45
    ="@
    -1.44
    "}](#
    -1.44
    junit
    -1.42
    же
    -1.38
    iasis
    -1.37
    blockList
    -1.35
    POSITIVE LOGITS
     souls
    1.67
    InstanceState
    1.62
     ownership
    1.60
    pass
    1.58
     lanes
    1.57
     sight
    1.57
     control
    1.56
    etine
    1.52
     possession
    1.47
     dB
    1.44
    Activations Density 0.149%

    No Known Activations