© 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. 4-RES-SM
    11. 9881
    Prev
    Next
    INDEX
    Explanations

    references to a specific context or environment, particularly "field" and "fields."

    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
    396
    +0.14
    0.8%
    9
    +0.11
    0.6%
    195
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.14
    0.02
    396
    +0.11
    0.02
    479
    +0.11
    0.01
    Negative Logits
    ¿
    -2.60
    ¦
    -2.42
    °
    -2.30
    ³
    -2.29
    ¾
    -2.26
    ¶
    -2.26
    ĻĤ
    -2.14
    «
    -2.13
    §
    -2.12
    Ń
    -2.09
    POSITIVE LOGITS
    work
    2.72
    workers
    2.29
    pieces
    2.04
    notes
    1.94
    piece
    1.85
    hog
    1.85
     Marshal
    1.85
    works
    1.81
    worker
    1.78
    names
    1.76
    Activations Density 0.160%

    No Known Activations