© 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. 12401
    Prev
    Next
    INDEX
    Explanations

    key terms related to environmental standards and educational policy

    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
    34
    +0.13
    0.7%
    9
    +0.12
    0.7%
    197
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    34
    +0.13
    0.12
    418
    +0.12
    0.01
    285
    +0.12
    0.08
    Negative Logits
     ich
    -1.72
    »
    -1.66
    ľĵ
    -1.61
     endless
    -1.53
    apshot
    -1.52
    ichever
    -1.49
    abbing
    -1.49
     unwanted
    -1.47
     mammalian
    -1.46
    ¯
    -1.45
    POSITIVE LOGITS
     Association
    2.48
     Authority
    2.37
     Committee
    2.30
     Branch
    2.29
     Organization
    2.28
     Management
    2.27
     Agency
    2.27
     Manager
    2.25
     Institute
    2.25
     Bureau
    2.22
    Activations Density 0.981%

    No Known Activations