© 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. 2-RES-SM
    11. 12779
    Prev
    Next
    INDEX
    Explanations

    the word "as" in various contexts

    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
    +0.14
    0.8%
    176
    +0.12
    0.7%
    91
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    344
    +0.14
    0.04
    91
    +0.12
    0.03
    176
    +0.11
    0.03
    Negative Logits
    §
    -2.49
    ĥ
    -2.26
    ¾
    -2.08
    ģ
    -2.07
    ij
    -2.04
    Īĺ
    -2.03
    Ľ
    -2.01
    ¨
    -1.96
    ŀ
    -1.87
    Ģ
    -1.80
    POSITIVE LOGITS
     others
    1.93
    others
    1.71
     else
    1.66
     other
    1.48
     elsewhere
    1.43
    googleapis
    1.41
    iddell
    1.36
     pardon
    1.35
     forgive
    1.32
    other
    1.29
    Activations Density 0.110%

    No Known Activations