© 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. 15014
    Prev
    Next
    INDEX
    Explanations

    elements and attributes in HTML or XML code

    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
    78
    +0.19
    1.2%
    21
    +0.16
    1.0%
    9
    +0.14
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    9
    +0.19
    0.05
    78
    +0.16
    0.01
    323
    +0.14
    0.04
    Negative Logits
     SY
    -1.33
     breaks
    -1.30
     mechanics
    -1.28
    estones
    -1.26
     designation
    -1.26
    atics
    -1.23
     favour
    -1.20
     coming
    -1.14
    ITCH
    -1.14
     franchise
    -1.13
    POSITIVE LOGITS
    ·¸
    3.42
    <
    3.39
    ¶
    3.35
    ķ
    3.30
    IJ
    3.26
    «
    3.21
    į
    3.21
    ·
    3.21
    ŀ
    3.19
    Ķ
    3.17
    Activations Density 0.203%

    No Known Activations