© 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. 0-MLP-SM
    6. 27743
    Prev
    Next
    INDEX
    Explanations

    occurrences of the SQL keyword "FROM"

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    83
    +0.12
    0.7%
    470
    +0.12
    0.7%
    346
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    352
    +0.12
    0.22
    185
    +0.12
    0.18
    165
    +0.12
    0.17
    Negative Logits
    ching
    -1.64
    king
    -1.63
    cule
    -1.56
    lyn
    -1.53
    enger
    -1.49
    oning
    -1.47
    ing
    -1.46
     runner
    -1.43
     watching
    -1.43
     contradiction
    -1.43
    POSITIVE LOGITS
    ĨĴ
    2.89
    ¿
    2.56
    Ĥ¬
    2.42
    ¦
    2.38
    ·¸
    2.35
    ij
    2.34
    ¤
    2.34
    Ģ
    2.31
    ¡
    2.25
    ¬
    2.14
    Activations Density 0.394%

    No Known Activations