© 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. 32209
    Prev
    Next
    INDEX
    Explanations

    references to professional experiences and roles within organizations

    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
    184
    +0.14
    0.8%
    282
    +0.12
    0.7%
    293
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    282
    +0.14
    0.11
    362
    +0.12
    0.11
    440
    +0.12
    0.07
    Negative Logits
    ricular
    -1.45
     boundary
    -1.42
    ensity
    -1.41
     signific
    -1.40
     glance
    -1.38
    ÃŃvel
    -1.37
     deform
    -1.37
    umab
    -1.36
     incidence
    -1.36
    åĢ¼
    -1.35
    POSITIVE LOGITS
    ľĵ
    3.45
    3.45
    <|outofrange|>
    3.45
    ↵  âĢĥ
    3.45
    ↵
    3.45
    ↵
    3.45
              ↵
    3.45
    č↵
    3.45
    3.45
    ↵
    3.45
    Activations Density 2.340%

    No Known Activations