© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 10086
    Prev
    Next
    INDEX
    Explanations

    sequences of symbols and formatting used in coding or technical documents

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    141
    +0.16
    0.9%
    248
    +0.13
    0.8%
    178
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    141
    +0.16
    0.12
    466
    +0.13
    0.13
    16
    +0.11
    0.11
    Negative Logits
    ĻĤ
    -2.41
    ¹
    -1.54
    ļ
    -1.54
     touches
    -1.50
    ħ
    -1.47
    nets
    -1.33
     stitches
    -1.32
     Ice
    -1.30
     CU
    -1.29
     Vill
    -1.26
    POSITIVE LOGITS
    gered
    1.55
    atically
    1.49
    rapeut
    1.45
    ynthe
    1.42
     liking
    1.40
     productivity
    1.38
     benchmark
    1.38
     LIABLE
    1.37
     dissemin
    1.35
    equivalent
    1.34
    Activations Density 6.425%

    No Known Activations