© 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. 2-MLP-SM
    11. 22403
    Prev
    Next
    INDEX
    Explanations

    repeated phrases or words that emphasize agreement or connection

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    201
    +0.14
    0.8%
    172
    +0.12
    0.6%
    97
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    97
    +0.14
    0.06
    338
    +0.12
    0.04
    78
    +0.12
    0.03
    Negative Logits
    nai
    -1.67
    )\].
    -1.59
    lyss
    -1.52
    psy
    -1.50
    peace
    -1.48
     wonders
    -1.48
    stick
    -1.47
    ivities
    -1.45
    mand
    -1.45
    \]).
    -1.44
    POSITIVE LOGITS
    ī
    3.53
    Ļª
    3.47
    Ī
    3.25
    ĩ
    3.24
    ģ
    3.22
    Ĵ
    3.21
    Ħ
    3.20
    Ĩ
    3.18
    Ĭ
    3.03
    ĭ
    3.02
    Activations Density 0.797%

    No Known Activations