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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 23939
    Prev
    Next
    INDEX
    Explanations

    phrases indicating past experiences or conditions

    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
    108
    +0.13
    0.7%
    310
    +0.12
    0.7%
    419
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    470
    +0.13
    0.05
    108
    +0.12
    0.05
    255
    +0.11
    0.05
    Negative Logits
    ished
    -1.62
    ]{.
    -1.59
     clinically
    -1.56
     precisely
    -1.55
    ths
    -1.50
     caution
    -1.50
     merely
    -1.47
    oned
    -1.45
    ted
    -1.45
     unanimously
    -1.44
    POSITIVE LOGITS
     &
    1.61
    starter
    1.48
    ())
    1.47
    oups
    1.44
    ás
    1.43
    ée
    1.42
    etc
    1.42
     etc
    1.41
    quote
    1.41
    icorn
    1.40
    Activations Density 0.271%

    No Known Activations