© 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. 23274
    Prev
    Next
    INDEX
    Explanations

    phrases related to training, motivation, and persistence

    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
    263
    +0.12
    0.7%
    327
    +0.11
    0.6%
    490
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    95
    +0.12
    0.13
    327
    +0.11
    0.15
    118
    +0.11
    0.16
    Negative Logits
    ħ
    -3.12
    ĻĤ
    -3.04
    ±
    -2.93
    ¾
    -2.61
    ¢
    -2.58
    ij
    -2.54
    ĥ
    -2.52
    ³
    -2.51
    »¿
    -2.50
    Ħ
    -2.47
    POSITIVE LOGITS
    allocate
    1.80
     optimized
    1.64
    initialize
    1.59
    inho
    1.52
     optimize
    1.51
    setlength
    1.50
     initialize
    1.50
    compile
    1.44
     differentiate
    1.41
    validate
    1.40
    Activations Density 4.165%

    No Known Activations