© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 25353
    Prev
    Next
    INDEX
    Explanations

    patterns or structures in programming languages or code snippets

    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
    +0.25
    1.4%
    413
    +0.11
    0.6%
    473
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.25
    0.13
    30
    +0.11
    0.04
    163
    +0.10
    0.02
    Negative Logits
    "."
    -1.72
    imum
    -1.68
    parency
    -1.54
    erred
    -1.49
     reason
    -1.44
    rito
    -1.42
    >",
    -1.42
     valuation
    -1.40
    ctrine
    -1.37
     sanity
    -1.36
    POSITIVE LOGITS
     together
    1.60
     papers
    1.50
     briefs
    1.47
     images
    1.45
     channels
    1.43
    nets
    1.43
     juices
    1.43
     supra
    1.43
     newspapers
    1.41
    prints
    1.40
    Activations Density 2.121%

    No Known Activations