© 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. 15308
    Prev
    Next
    INDEX
    Explanations

    specific technical parameters and configurations related to software components

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    357
    +0.18
    1.1%
    11
    +0.16
    1.0%
    203
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    357
    +0.18
    0.12
    203
    +0.16
    0.09
    397
    +0.14
    0.10
    Negative Logits
    icago
    -1.55
    uesday
    -1.41
    ugu
    -1.37
    watson
    -1.34
    john
    -1.32
    ```
    -1.31
    ski
    -1.31
    idis
    -1.31
     americ
    -1.30
    alex
    -1.29
    POSITIVE LOGITS
     Value
    2.29
     Calendar
    2.25
     Values
    2.24
     Events
    2.21
     Event
    2.12
     Purpose
    2.11
     Control
    2.09
     Device
    2.08
     Activity
    2.07
     Effect
    2.06
    Activations Density 1.885%

    No Known Activations