© 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. 31061
    Prev
    Next
    INDEX
    Explanations

    numeric values represented in different forms or contexts

    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
    100
    +0.18
    1.0%
    478
    +0.17
    1.0%
    393
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.18
    0.03
    100
    +0.17
    0.03
    165
    +0.12
    0.02
    Negative Logits
    ļ
    -1.58
    icago
    -1.45
    ialog
    -1.41
     well
    -1.37
     makers
    -1.36
     advantage
    -1.35
    HECK
    -1.34
     MW
    -1.33
    isp
    -1.33
    àµį
    -1.33
    POSITIVE LOGITS
    :]
    1.51
    occupied
    1.43
    —.
    1.43
    TRODUCTION
    1.43
    ável
    1.35
    ¼
    1.33
    "_
    1.32
    raged
    1.32
    bre
    1.31
     parole
    1.27
    Activations Density 0.078%

    No Known Activations