© 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. 25905
    Prev
    Next
    INDEX
    Explanations

    references to fictional or imagined entities

    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
    23
    +0.15
    0.9%
    364
    +0.12
    0.7%
    45
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.15
    0.02
    339
    +0.12
    0.02
    341
    +0.11
    0.01
    Negative Logits
     workshops
    -1.59
    imos
    -1.56
    posium
    -1.52
    ographies
    -1.51
    ÃŃf
    -1.50
     basics
    -1.48
    ÃŃs
    -1.45
     remedies
    -1.44
     rendition
    -1.43
    icro
    -1.43
    POSITIVE LOGITS
     faint
    1.69
    lay
    1.68
    stra
    1.54
    lington
    1.36
     Line
    1.33
    ANI
    1.33
    abled
    1.32
    pike
    1.31
    king
    1.31
    unov
    1.31
    Activations Density 0.042%

    No Known Activations