© 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. 2-RES-SM
    11. 13560
    Prev
    Next
    INDEX
    Explanations

    terms related to medical contexts and conditions

    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
    18
    +0.16
    0.9%
    250
    +0.15
    0.9%
    391
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    18
    +0.16
    0.03
    391
    +0.15
    0.02
    250
    +0.14
    0.03
    Negative Logits
    hift
    -1.64
    olated
    -1.56
    oming
    -1.45
    ses
    -1.41
    ctions
    -1.40
    reptococcus
    -1.37
    âĭ¯
    -1.31
    akes
    -1.30
     tires
    -1.30
     saving
    -1.27
    POSITIVE LOGITS
    iator
    2.59
    ¾
    2.15
    iates
    2.12
    iterr
    2.03
    dle
    2.03
    itation
    2.02
    ieval
    2.00
    iterranean
    1.97
    ¿
    1.94
    ieux
    1.86
    Activations Density 0.159%

    No Known Activations