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

    Neuronpedia

    APICircuit 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. 11556
    Prev
    Next
    INDEX
    Explanations

    technical terms and terminology related to programming or software development

    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
    271
    +0.18
    1.0%
    263
    +0.16
    0.9%
    72
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.18
    -0.02
    204
    +0.16
    0.11
    498
    +0.16
    0.10
    Negative Logits
    unnumbered
    -1.52
    CLA
    -1.52
     Species
    -1.49
     blacks
    -1.37
    onitrile
    -1.35
    poons
    -1.34
     Poor
    -1.32
    olation
    -1.32
     Census
    -1.31
    UTH
    -1.29
    POSITIVE LOGITS
    ¼
    3.11
    ¨
    2.87
    Ļ
    2.84
    º
    2.80
    ®
    2.77
    Ń
    2.64
    ¢
    2.62
    «
    2.62
    ł
    2.58
    Ł
    2.58
    Activations Density 2.536%

    No Known Activations