© 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. 22980
    Prev
    Next
    INDEX
    Explanations

    occurrences of the phrases "one or more."

    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
    261
    +0.13
    0.7%
    74
    +0.12
    0.7%
    298
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    10
    +0.13
    0.04
    261
    +0.12
    0.02
    393
    +0.11
    0.03
    Negative Logits
    uthan
    -1.71
    aptic
    -1.67
    ugin
    -1.50
     praying
    -1.40
    edia
    -1.39
    smallcaps
    -1.38
     matters
    -1.37
    dala
    -1.35
     receivers
    -1.35
    ipl
    -1.33
    POSITIVE LOGITS
    ¬
    2.03
    »
    2.02
    ī
    1.99
    Ī
    1.90
    İ
    1.88
    ŀ
    1.87
    naments
    1.83
    §
    1.82
    Ĵ
    1.70
    ģ
    1.69
    Activations Density 0.097%

    No Known Activations