© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 26959
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "the."

    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/2-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.2.hook_mlp_out
    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
    54
    +0.14
    0.8%
    92
    +0.13
    0.7%
    22
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    92
    +0.14
    0.07
    22
    +0.13
    0.06
    291
    +0.11
    0.07
    Negative Logits
    ³
    -2.54
    ī
    -2.18
    ¯
    -1.96
    ¿
    -1.95
    ´
    -1.92
    ²
    -1.90
    ¿½
    -1.89
    §
    -1.86
    Ĵ
    -1.85
    »
    -1.83
    POSITIVE LOGITS
     alike
    1.81
    mate
    1.61
     lives
    1.53
     detriment
    1.49
    occup
    1.47
    seeking
    1.46
    istically
    1.45
    manship
    1.42
    heed
    1.42
     times
    1.41
    Activations Density 0.640%

    No Known Activations