Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 12101
    Prev
    Next
    INDEX
    Explanations

    mentions of specific locations 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
    478
    +0.17
    1.0%
    489
    +0.12
    0.7%
    331
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    478
    +0.17
    0.00
    124
    +0.12
    0.06
    331
    +0.10
    0.06
    Negative Logits
    ŀ
    -1.90
    ĩ
    -1.86
    ľĵ
    -1.86
    ľ
    -1.85
     themselves
    -1.73
    Ļª
    -1.67
    Ŀ
    -1.63
    £
    -1.54
    ļ
    -1.54
    ĸ
    -1.49
    POSITIVE LOGITS
    lining
    1.56
     intention
    1.40
    aid
    1.39
     uniqueness
    1.38
     NOTICE
    1.36
     hypocr
    1.35
    ported
    1.34
     behalf
    1.33
     workflow
    1.32
    paired
    1.31
    Activations Density 2.299%

    No Known Activations