© 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. 23380
    Prev
    Next
    INDEX
    Explanations

    the presence of the substring "pl" across 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
    +0.17
    0.9%
    326
    +0.14
    0.8%
    290
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.17
    0.01
    374
    +0.14
    0.01
    8
    +0.11
    0.02
    Negative Logits
    hoe
    -1.94
    blr
    -1.86
    googleapis
    -1.78
    oxford
    -1.67
    comed
    -1.62
    npmjs
    -1.60
    adoc
    -1.58
    orems
    -1.56
    gom
    -1.55
    atham
    -1.54
    POSITIVE LOGITS
    µ
    2.05
    °
    2.04
    ´
    1.98
    Ń
    1.88
    ŀ
    1.82
    ±
    1.75
    ł
    1.74
    Ĭ
    1.73
    ģ
    1.64
    ¦
    1.63
    Activations Density 0.234%

    No Known Activations