© 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. 28901
    Prev
    Next
    INDEX
    Explanations

    references to legal districts and court cases

    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
    111
    +0.15
    0.9%
    248
    +0.15
    0.8%
    263
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    319
    +0.15
    0.02
    74
    +0.15
    0.02
    34
    +0.11
    0.03
    Negative Logits
     respectively
    -1.55
     whilst
    -1.52
     additive
    -1.48
     traditional
    -1.48
     sensor
    -1.37
     Australians
    -1.37
     saving
    -1.35
     realise
    -1.31
    plastic
    -1.31
     conventional
    -1.31
    POSITIVE LOGITS
    ¸
    2.99
    ĭ
    2.56
    Ģ
    2.51
    Ļª
    2.44
    ī
    2.43
    Ī
    2.39
    Īĺ
    2.37
    ½
    2.34
    ¹
    2.32
    ¤
    2.32
    Activations Density 0.173%

    No Known Activations