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. 11829
    Prev
    Next
    INDEX
    Explanations

    instances of significant numbers and their implications

    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
    203
    +0.15
    0.9%
    478
    +0.13
    0.8%
    136
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.15
    0.11
    136
    +0.13
    -0.01
    23
    +0.13
    0.11
    Negative Logits
    ceral
    -1.89
    anguage
    -1.86
    iveness
    -1.63
    ,
    -1.52
    ings
    -1.51
    -âĤ¬
    -1.51
    STA
    -1.50
    iously
    -1.48
    ifice
    -1.44
    ģ
    -1.41
    POSITIVE LOGITS
    which
    1.68
    000
    1.56
    omitempty
    1.51
     hence
    1.47
    Ïģγ
    1.43
     son
    1.41
     nor
    1.41
     któ
    1.40
    ко
    1.40
    lem
    1.39
    Activations Density 1.057%

    No Known Activations