© 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. 27110
    Prev
    Next
    INDEX
    Explanations

    instances of the opening square bracket character

    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
    188
    +0.20
    1.1%
    410
    +0.15
    0.8%
    478
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    188
    +0.20
    0.01
    51
    +0.15
    0.00
    424
    +0.14
    0.01
    Negative Logits
    osten
    -1.67
    omed
    -1.51
    falls
    -1.46
    inar
    -1.46
    äsident
    -1.43
    olev
    -1.40
    urse
    -1.40
    ikel
    -1.40
    itates
    -1.40
     supernatants
    -1.39
    POSITIVE LOGITS
    unreadable
    1.87
    ...)
    1.76
    PUBL
    1.63
    *)
    1.53
    thick
    1.52
    pard
    1.49
    illion
    1.49
    ...]
    1.49
    inher
    1.48
    )$-
    1.48
    Activations Density 0.018%

    No Known Activations