© 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. 4-RES-SM
    11. 8920
    Prev
    Next
    INDEX
    Explanations

    the word "among" and its variations, indicating a focus on group or collective 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/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
    307
    +0.24
    1.5%
    306
    +0.18
    1.1%
    466
    +0.16
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    306
    +0.24
    0.02
    98
    +0.18
    0.01
    307
    +0.16
    0.03
    Negative Logits
    hip
    -1.78
    isation
    -1.52
    ridge
    -1.50
    ledge
    -1.40
    itrile
    -1.34
    vation
    -1.33
    iosis
    -1.31
    ivery
    -1.29
     relationship
    -1.28
    ving
    -1.28
    POSITIVE LOGITS
     others
    2.63
    them
    2.20
    others
    2.13
    ĨĴ
    2.08
    ĻĤ
    2.07
     them
    2.06
     other
    2.02
    ĥ½
    1.96
    Īĺ
    1.82
    į
    1.73
    Activations Density 0.209%

    No Known Activations