© 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. 25392
    Prev
    Next
    INDEX
    Explanations

    occurrences of access modifiers like "public" and "private" in programming code

    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
    122
    +0.12
    0.7%
    40
    +0.12
    0.7%
    391
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    166
    +0.12
    0.06
    122
    +0.12
    0.06
    134
    +0.12
    0.07
    Negative Logits
     engineers
    -1.65
     operators
    -1.63
     \%
    -1.55
     differential
    -1.55
     journals
    -1.51
     exemplary
    -1.50
     readers
    -1.48
     respondents
    -1.47
     papers
    -1.46
     cart
    -1.46
    POSITIVE LOGITS
    ĨĴ
    1.98
    etto
    1.91
    Īĺ
    1.76
    garten
    1.67
    ľĵ
    1.66
    imum
    1.64
    ·
    1.64
    isman
    1.62
    iento
    1.62
    disambiguation
    1.60
    Activations Density 0.352%

    No Known Activations