© 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. 23645
    Prev
    Next
    INDEX
    Explanations

    instances of the word "important" and its variations, indicating a focus on significant issues or concepts

    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
    496
    +0.13
    0.7%
    387
    +0.12
    0.6%
    324
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.13
    0.07
    179
    +0.12
    0.07
    306
    +0.11
    0.06
    Negative Logits
    ±
    -2.98
    ª
    -2.93
    ³
    -2.81
    ¸
    -2.76
    ķ
    -2.67
    ¦
    -2.58
    ł
    -2.49
    ²
    -2.47
    ¹
    -2.42
    IJ
    -2.42
    POSITIVE LOGITS
    blank
    1.67
    Ãħ
    1.66
    uthan
    1.42
    ,&
    1.42
    gie
    1.38
     Dise
    1.37
    /,
    1.37
    /-
    1.33
     blank
    1.33
     BASIS
    1.32
    Activations Density 0.975%

    No Known Activations