© 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. 22860
    Prev
    Next
    INDEX
    Explanations

    instances and mentions of "um" in the context of conversation or writing

    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
    148
    +0.14
    0.7%
    23
    +0.13
    0.7%
    39
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.14
    0.02
    374
    +0.13
    0.01
    232
    +0.13
    0.03
    Negative Logits
    į
    -2.10
    americ
    -1.77
    ĵ
    -1.73
    Ĭ
    -1.66
    mente
    -1.62
    Ģ
    -1.58
    ı
    -1.56
    cdn
    -1.55
    adult
    -1.49
    rugs
    -1.49
    POSITIVE LOGITS
     plenty
    1.45
     happening
    1.44
     pile
    1.40
     \'
    1.37
     congestion
    1.37
     crunch
    1.36
     thrown
    1.36
    ?’
    1.35
     diligence
    1.33
     ming
    1.32
    Activations Density 0.341%

    No Known Activations