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. 11915
    Prev
    Next
    INDEX
    Explanations

    terms related to software licensing and file organization

    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
    198
    +0.17
    0.9%
    111
    +0.12
    0.7%
    47
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    491
    +0.17
    0.04
    11
    +0.12
    0.02
    107
    +0.12
    0.02
    Negative Logits
    []{
    -1.77
    ĥ½
    -1.71
     []{
    -1.52
    Ł
    -1.44
     respectively
    -1.43
    respectively
    -1.42
    ...](
    -1.40
    Ĺ
    -1.36
    decoration
    -1.36
     ↵↵
    -1.34
    POSITIVE LOGITS
    )—
    1.76
    tico
    1.62
    usiness
    1.56
    othes
    1.51
    )"
    1.47
    liness
    1.45
    atory
    1.41
    ously
    1.38
    laws
    1.35
    ible
    1.34
    Activations Density 0.245%

    No Known Activations