© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 30926
    Prev
    Next
    INDEX
    Explanations

    repeated use of the pronoun "we" to emphasize collective action or perspective

    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
    421
    +0.15
    0.8%
    122
    +0.13
    0.7%
    202
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.15
    0.10
    356
    +0.13
    0.11
    498
    +0.13
    0.10
    Negative Logits
    ¢
    -1.78
    ¨
    -1.67
    escence
    -1.66
    imus
    -1.58
    §
    -1.54
    izing
    -1.53
    ¦
    -1.47
    egal
    -1.39
    ĻĤ
    -1.38
     elected
    -1.38
    POSITIVE LOGITS
    orms
    1.60
     Actions
    1.41
    _[
    1.41
    load
    1.40
     DAMAGE
    1.39
    ças
    1.39
    ppings
    1.37
    atures
    1.35
    occur
    1.34
    illon
    1.32
    Activations Density 0.255%

    No Known Activations