© 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. 14122
    Prev
    Next
    INDEX
    Explanations

    prepositions and infinitive forms of verbs

    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
    478
    +0.21
    1.2%
    69
    +0.15
    0.9%
    220
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.21
    0.14
    434
    +0.15
    0.10
    300
    +0.12
    0.08
    Negative Logits
     aside
    -1.95
     enough
    -1.71
    stitutional
    -1.65
    nes
    -1.58
    isco
    -1.53
    sense
    -1.53
     thought
    -1.50
    cap
    -1.49
     objection
    -1.47
     irony
    -1.45
    POSITIVE LOGITS
    Īĺ
    2.18
    ĥ
    2.18
    ©
    2.10
    ¥
    2.08
    asts
    1.97
    ķ
    1.95
    ¿½
    1.94
    ↵  âĢĥ
    1.81
    1.81
    ↵
    1.81
    Activations Density 0.638%

    No Known Activations