© 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. 10769
    Prev
    Next
    INDEX
    Explanations

    concepts related to life circumstances and choices

    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
    326
    +0.18
    1.1%
    17
    +0.18
    1.0%
    197
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.18
    0.06
    56
    +0.18
    -0.02
    328
    +0.14
    0.07
    Negative Logits
    »
    -2.48
    ¦
    -2.13
    ©
    -2.07
    ³
    -2.07
    ↵↵
    -2.07
    <|outofrange|>
    -2.07
    ↵
    -2.07
     ↵
    -2.07
    -2.07
               ↵
    -2.07
    POSITIVE LOGITS
    comings
    1.53
    ?_
    1.49
     paradise
    1.49
     kingdom
    1.49
     pleasures
    1.48
     sanct
    1.42
    boat
    1.41
     saf
    1.39
    hab
    1.38
     occupant
    1.37
    Activations Density 2.470%

    No Known Activations