© 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. 12489
    Prev
    Next
    INDEX
    Explanations

    phrases related to terms and conditions or legal rights

    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
    419
    +0.15
    0.9%
    69
    +0.14
    0.8%
    193
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.15
    0.05
    193
    +0.14
    0.04
    358
    +0.13
    0.04
    Negative Logits
    ĻĤ
    -4.45
    ĸ
    -3.87
    ĨĴ
    -3.85
    Īĺ
    -3.85
    ¤
    -3.82
    ª
    -3.81
    ¿½
    -3.72
    Ļª
    -3.65
    Ĵ
    -3.61
    ķ
    -3.57
    POSITIVE LOGITS
    ancell
    1.71
     judicial
    1.60
     oath
    1.58
     publication
    1.55
    unds
    1.49
    ating
    1.47
     affirmative
    1.46
     entirety
    1.46
     repr
    1.46
     civil
    1.45
    Activations Density 0.355%

    No Known Activations