© 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. 11084
    Prev
    Next
    INDEX
    Explanations

    actions related to the production, release, and establishment of various media and initiatives

    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
    21
    +0.15
    0.8%
    15
    +0.11
    0.6%
    184
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.15
    0.06
    15
    +0.11
    0.07
    463
    +0.10
    0.04
    Negative Logits
    »¿
    -2.85
    ĥ½
    -2.74
    č↵
    -2.52
     ↵
    -2.52
              ↵
    -2.52
    ↵
    -2.52
    ↵
    -2.52
    -2.52
     ↵
    -2.52
    ↵
    -2.52
    POSITIVE LOGITS
     thanks
    1.76
     unanimously
    1.74
     himself
    1.65
     farewell
    1.49
     late
    1.42
     adj
    1.38
     redesign
    1.34
     yesterday
    1.34
     Thursday
    1.34
     stern
    1.33
    Activations Density 0.411%

    No Known Activations