© 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. 4-RES-SM
    11. 10052
    Prev
    Next
    INDEX
    Explanations

    instances of the phrase "least 9" or similar numerical thresholds

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    25
    +0.15
    0.8%
    21
    +0.12
    0.7%
    261
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.15
    0.03
    373
    +0.12
    0.03
    121
    +0.11
    0.02
    Negative Logits
    ½
    -3.41
    ·¸
    -3.37
    ¼
    -3.36
    ´
    -3.28
    ¿
    -3.24
    ²
    -3.16
    ľĵ
    -3.08
    ĸ´
    -3.07
    ®
    -3.00
    ©
    -2.97
    POSITIVE LOGITS
     one
    2.15
     two
    2.07
     four
    1.97
     three
    1.96
     six
    1.91
     five
    1.88
     half
    1.78
     twice
    1.71
     partially
    1.70
    half
    1.70
    Activations Density 0.147%

    No Known Activations