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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 2-MLP-SM
    6. 25875
    Prev
    Next
    INDEX
    Explanations

    phrases related to programming concepts and structures

    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
    416
    +0.14
    0.8%
    429
    +0.13
    0.7%
    131
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    92
    +0.14
    0.03
    59
    +0.13
    -0.01
    286
    +0.12
    0.02
    Negative Logits
     %.
    -1.57
    rvert
    -1.41
    unct
    -1.35
     integers
    -1.34
    kan
    -1.28
    otoxin
    -1.27
    rVert
    -1.26
     heights
    -1.26
    fileID
    -1.26
    posted
    -1.25
    POSITIVE LOGITS
     authorities
    1.57
     spokesman
    1.56
    ifier
    1.49
     importantly
    1.44
    roller
    1.44
     pursued
    1.41
    ificates
    1.41
     finally
    1.39
    "?
    1.37
    "?"
    1.36
    Activations Density 1.191%

    No Known Activations