© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 4-RES-SM
    6. 8658
    Prev
    Next
    INDEX
    Explanations

    references to programming concepts and syntax structures

    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
    156
    +0.40
    2.7%
    397
    +0.18
    1.3%
    136
    +0.18
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    397
    +0.40
    0.12
    21
    +0.18
    0.04
    9
    +0.18
    0.07
    Negative Logits
     yours
    -1.27
     (,
    -1.26
     works
    -1.21
    selves
    -1.19
    ”:
    -1.18
     apparatuses
    -1.17
    estly
    -1.15
    iously
    -1.14
    ess
    -1.13
     companies
    -1.12
    POSITIVE LOGITS
    )));
    1.50
    )),
    1.46
    ]));
    1.39
    ]\]).
    1.33
    ))))
    1.33
     respectively
    1.33
    astolic
    1.30
    )].
    1.29
    )).
    1.29
     predecessor
    1.24
    Activations Density 6.119%

    No Known Activations