© 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 EvalsExportsAPI 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. 24846
    Prev
    Next
    INDEX
    Explanations

    instances of the word "swirl" in various contexts

    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
    156
    +0.18
    1.0%
    367
    +0.12
    0.7%
    31
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    367
    +0.18
    0.01
    42
    +0.12
    0.04
    491
    +0.11
    0.02
    Negative Logits
     hers
    -1.65
    ĥ½
    -1.63
    mium
    -1.58
     ours
    -1.56
     loved
    -1.46
    º
    -1.43
     theirs
    -1.39
    "){
    -1.35
    %"
    -1.34
    ugu
    -1.33
    POSITIVE LOGITS
    ulous
    1.54
    itas
    1.42
    ated
    1.42
    |
    1.40
    hbar
    1.38
    gment
    1.38
    atted
    1.36
    angling
    1.36
     Order
    1.34
    anking
    1.33
    Activations Density 0.422%

    No Known Activations