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. MLP Post
    5. 0-MLP-SM
    6. 27296
    Prev
    Next
    INDEX
    Explanations

    instances of the term "substitute" or its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    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
    376
    +0.17
    1.0%
    59
    +0.14
    0.9%
    40
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    59
    +0.17
    0.01
    40
    +0.14
    0.01
    15
    +0.14
    0.01
    Negative Logits
    ³
    -4.37
    ¿½
    -4.35
    ĨĴ
    -4.28
    ĥ½
    -4.22
    »¿
    -4.20
    Ī
    -4.20
    ¶
    -4.19
    Ļª
    -4.19
    ľ
    -4.13
    ²
    -4.10
    POSITIVE LOGITS
    etine
    2.23
    atics
    1.96
    ário
    1.89
    quist
    1.83
    eto
    1.80
    hdad
    1.78
    hold
    1.78
    iors
    1.76
     positions
    1.74
    ematic
    1.68
    Activations Density 4.667%

    No Known Activations