© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 28212
    Prev
    Next
    INDEX
    Explanations

    programming-related syntax and patterns in the 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/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
    138
    +0.12
    0.7%
    218
    +0.11
    0.6%
    312
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    322
    +0.12
    -0.01
    74
    +0.11
    0.06
    431
    +0.10
    0.05
    Negative Logits
    eson
    -1.66
    Ļª
    -1.62
    icable
    -1.61
    »¿
    -1.53
    ullary
    -1.48
    existent
    -1.46
    InstanceState
    -1.41
    wordpress
    -1.40
    eed
    -1.38
    rencies
    -1.38
    POSITIVE LOGITS
     newsletter
    1.44
    ^](#
    1.44
    /-
    1.38
     haste
    1.26
    edom
    1.25
     plus
    1.24
    lining
    1.24
     auditor
    1.23
    lift
    1.22
     »
    1.21
    Activations Density 5.253%

    No Known Activations