© 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. 1-RES-SM
    11. 25326
    Prev
    Next
    INDEX
    Explanations

    the presence of the letter 'N' or similar characters in text

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    310
    +0.14
    0.8%
    264
    +0.12
    0.6%
    267
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    429
    +0.14
    0.00
    18
    +0.12
    0.00
    272
    +0.11
    0.00
    Negative Logits
    ¼
    -1.75
    abouts
    -1.67
    ĭ
    -1.67
    Ł
    -1.63
    )).
    -1.59
    İ
    -1.58
    });
    -1.57
    aves
    -1.55
    )),
    -1.50
    ĻĤ
    -1.49
    POSITIVE LOGITS
    fully
    1.55
    ened
    1.48
    amycin
    1.47
    insula
    1.45
    ||
    1.44
    falls
    1.38
     extent
    1.36
    patient
    1.36
    pez
    1.33
    IPE
    1.33
    Activations Density 0.001%

    No Known Activations