© 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. 22856
    Prev
    Next
    INDEX
    Explanations

    structure or elements in a document that indicate the presence of lists or items

    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
    156
    +0.27
    1.5%
    261
    +0.12
    0.7%
    226
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    261
    +0.27
    0.01
    50
    +0.12
    0.01
    210
    +0.11
    0.00
    Negative Logits
    ¡
    -2.42
    IJ
    -2.35
    Ł
    -2.29
    ¨
    -2.28
    Ĺ
    -2.17
    Ļª
    -2.12
    Ĵ
    -2.04
    ¦
    -2.01
    č↵
    -2.01
    -2.01
    POSITIVE LOGITS
    owitz
    1.78
    edly
    1.64
    ess
    1.54
    zza
    1.48
    emen
    1.48
    cum
    1.46
    elike
    1.46
    eding
    1.44
    uously
    1.38
    emis
    1.37
    Activations Density 0.122%

    No Known Activations