© 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. 22675
    Prev
    Next
    INDEX
    Explanations

    phrases related to technological features and specifications

    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
    481
    +0.13
    0.8%
    394
    +0.12
    0.7%
    73
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    118
    +0.13
    0.18
    142
    +0.12
    0.16
    362
    +0.10
    0.10
    Negative Logits
     learn
    -1.52
    iser
    -1.51
     conceive
    -1.44
     associate
    -1.43
    iates
    -1.37
     distinguish
    -1.37
     "
    -1.35
     _"
    -1.35
     marrow
    -1.35
     neut
    -1.35
    POSITIVE LOGITS
    :`
    1.67
    ’
    1.47
    ’,
    1.41
    blic
    1.41
    oto
    1.36
     plays
    1.33
    blr
    1.31
    й
    1.31
    thead
    1.29
    $’
    1.28
    Activations Density 2.384%

    No Known Activations