© 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. 0-MLP-SM
    11. 26818
    Prev
    Next
    INDEX
    Explanations

    instances of potentially grammatical items, such as proper nouns or names

    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
    239
    +0.14
    0.8%
    28
    +0.13
    0.7%
    151
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    438
    +0.14
    0.40
    53
    +0.13
    0.42
    56
    +0.10
    0.11
    Negative Logits
    ·¸
    -2.03
    º
    -1.95
    ª
    -1.67
    ¹
    -1.64
    µ
    -1.63
    ·
    -1.61
    ĭ
    -1.57
    alle
    -1.52
    ¨
    -1.49
    ¾
    -1.49
    POSITIVE LOGITS
    \]](
    1.85
     alike
    1.61
    ]{.
    1.57
    prising
    1.56
    .](
    1.47
    prises
    1.41
     clicked
    1.35
    proving
    1.35
     levels
    1.34
    itzer
    1.34
    Activations Density 4.634%

    No Known Activations