© 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 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. 24352
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate movement or transformation

    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
    242
    +0.12
    0.6%
    271
    +0.11
    0.6%
    14
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    100
    +0.12
    0.06
    70
    +0.11
    0.04
    14
    +0.10
    0.05
    Negative Logits
    Ļª
    -3.25
    ģ
    -2.74
    ĻĤ
    -2.71
    Ģ
    -2.69
    Ĩ
    -2.68
    ı
    -2.66
    ĥ
    -2.62
    ĨĴ
    -2.59
    ¿½
    -2.59
    Ĥ¬
    -2.57
    POSITIVE LOGITS
    ário
    1.70
    anas
    1.70
    ibase
    1.67
    isite
    1.55
    ık
    1.55
    ents
    1.52
    ese
    1.51
    oise
    1.51
    acity
    1.46
    ental
    1.46
    Activations Density 0.138%

    No Known Activations