© 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. 30888
    Prev
    Next
    INDEX
    Explanations

    statements made by individuals, particularly in formal 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
    478
    +0.15
    0.9%
    280
    +0.13
    0.7%
    392
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    372
    +0.15
    0.04
    462
    +0.13
    0.04
    41
    +0.11
    0.06
    Negative Logits
    »¿
    -2.55
    °
    -2.41
    ·¸
    -2.41
    ¬
    -2.29
    ħ
    -2.23
    ĭ
    -2.21
    Ļª
    -2.20
    Īĺ
    -2.19
    ķ
    -2.11
    ¹
    -2.05
    POSITIVE LOGITS
     goodbye
    1.69
    ward
    1.62
     blunt
    1.50
     falsely
    1.45
    appropri
    1.42
     quietly
    1.39
     wrongly
    1.37
     helpless
    1.34
     incorrectly
    1.34
     farewell
    1.32
    Activations Density 0.404%

    No Known Activations