© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 28728
    Prev
    Next
    INDEX
    Explanations

    references and citations in academic texts

    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
    282
    +0.14
    0.8%
    136
    +0.13
    0.7%
    478
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    136
    +0.14
    -0.00
    302
    +0.13
    0.02
    243
    +0.12
    0.03
    Negative Logits
    ħ
    -2.66
    Ģ
    -2.42
    ĭ
    -2.27
    Ļª
    -2.21
    Ļ
    -2.18
    ĩ
    -2.18
    ŀ
    -2.16
    Ħ
    -2.15
    ¬
    -2.13
    ģ
    -2.12
    POSITIVE LOGITS
    ){#
    1.82
     Hence
    1.78
     However
    1.68
     Therefore
    1.68
     Besides
    1.67
    However
    1.63
    since
    1.56
    Nevertheless
    1.55
     Thus
    1.54
    Hence
    1.54
    Activations Density 0.153%

    No Known Activations