© 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. 23484
    Prev
    Next
    INDEX
    Explanations

    instances of the word "John" and variations of "No" as well as expressions emphasizing the concept of negation or disagreement

    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
    111
    +0.49
    3.6%
    156
    +0.48
    3.5%
    137
    +0.09
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    497
    +0.49
    0.00
    414
    +0.48
    0.00
    275
    +0.09
    0.00
    Negative Logits
    ÑĢав
    -1.42
     reasonable
    -1.31
    ÑĢоÑģ
    -1.28
    нÑĭ
    -1.24
    asm
    -1.24
    1451450014514500
    -1.23
    lVert
    -1.23
     genuine
    -1.22
    ighters
    -1.21
     grateful
    -1.20
    POSITIVE LOGITS
     liner
    1.40
    arner
    1.25
     snap
    1.23
    istor
    1.23
     snaps
    1.16
     lineup
    1.16
     slider
    1.15
     debates
    1.15
     rock
    1.12
     voting
    1.12
    Activations Density 0.029%

    No Known Activations