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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 28367
    Prev
    Next
    INDEX
    Explanations

    phrases related to significant chances or potential for advancement

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    38
    +0.15
    0.8%
    166
    +0.11
    0.6%
    376
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    38
    +0.15
    0.04
    166
    +0.11
    0.03
    473
    +0.11
    0.02
    Negative Logits
    Ń
    -3.38
    Ĥ
    -3.38
    Ľ
    -3.14
    ĥ½
    -3.12
    ĨĴ
    -3.04
    °
    -3.02
    ĩ
    -3.01
    ¸
    -2.98
    ³
    -2.96
    ¯
    -2.94
    POSITIVE LOGITS
    ously
    1.78
     opportunities
    1.72
     opportunity
    1.71
     seekers
    1.69
    ous
    1.68
    holder
    1.67
    ground
    1.67
    grounds
    1.67
    ality
    1.60
    hole
    1.57
    Activations Density 0.129%

    No Known Activations