© 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. 23363
    Prev
    Next
    INDEX
    Explanations

    occurrences of class declarations in programming code

    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
    329
    +0.12
    0.6%
    364
    +0.11
    0.6%
    462
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.12
    0.03
    456
    +0.11
    0.01
    462
    +0.10
    0.02
    Negative Logits
    sworth
    -1.51
     0000000000000000000000000000000000
    -1.49
    ancer
    -1.42
    ancy
    -1.42
    áĢ¬
    -1.39
    imental
    -1.35
    ighting
    -1.33
     pride
    -1.33
     hell
    -1.33
    antically
    -1.29
    POSITIVE LOGITS
    º
    2.02
    ¹
    1.93
    Ī
    1.93
    ±
    1.81
    Ĵ
    1.81
    ŀ
    1.75
    ¼
    1.75
    Ĥ¬
    1.74
    Īĺ
    1.70
    ¢
    1.61
    Activations Density 0.053%

    No Known Activations