© 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 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. 11373
    Prev
    Next
    INDEX
    Explanations

    prepositions and words related to direction or purpose

    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
    412
    +0.12
    0.7%
    69
    +0.12
    0.6%
    245
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.12
    0.07
    245
    +0.12
    0.06
    16
    +0.11
    0.04
    Negative Logits
    teen
    -1.68
    ologists
    -1.45
     enough
    -1.41
     rays
    -1.41
     earthquakes
    -1.39
    ators
    -1.39
    vier
    -1.39
    wise
    -1.36
    shire
    -1.35
    orious
    -1.35
    POSITIVE LOGITS
    ·¸
    2.13
    ¥
    2.11
    ³
    2.08
    Ł
    2.00
    Ļª
    1.99
    ķ
    1.99
    IJ
    1.98
    Ķ
    1.97
    »¿
    1.95
    »
    1.89
    Activations Density 0.447%

    No Known Activations