© 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. 0-MLP-SM
    11. 29428
    Prev
    Next
    INDEX
    Explanations

    the word "celebrated" and its variations, indicating themes of recognition and acknowledgment

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    +0.22
    1.3%
    376
    +0.22
    1.3%
    148
    +0.19
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    215
    +0.22
    0.01
    148
    +0.22
    0.01
    415
    +0.19
    0.01
    Negative Logits
    IJ
    -2.33
    ħ
    -2.24
    į
    -2.07
    ij
    -1.97
    Ļ
    -1.93
    §
    -1.87
    İ
    -1.86
    ĥ½
    -1.85
    ĺ
    -1.82
    Ŀ
    -1.81
    POSITIVE LOGITS
    uate
    1.63
     noises
    1.46
    ding
    1.45
    ments
    1.41
    itious
    1.38
    asting
    1.38
    ational
    1.36
    [^
    1.36
     moments
    1.35
    ités
    1.34
    Activations Density 0.010%

    No Known Activations