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. 11907
    Prev
    Next
    INDEX
    Explanations

    numerical data and parameters related to data handling or programming 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
    161
    +0.19
    1.1%
    369
    +0.16
    1.0%
    17
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    161
    +0.19
    0.07
    143
    +0.16
    0.09
    121
    +0.15
    0.09
    Negative Logits
     (@
    -1.51
    )."
    -1.38
    ."[
    -1.33
    ainer
    -1.31
    icator
    -1.30
    ivist
    -1.29
    .^[@
    -1.27
    ichte
    -1.27
    ]{}]{}
    -1.26
    ale
    -1.26
    POSITIVE LOGITS
    ĥ½
    2.07
    ĸ´
    1.94
    ģ
    1.83
    ī
    1.71
    »¿
    1.71
    ¤
    1.68
    Ĥ¬
    1.64
    Ĺ
    1.64
    ±
    1.62
    ¾
    1.55
    Activations Density 4.871%

    No Known Activations