© 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. 22683
    Prev
    Next
    INDEX
    Explanations

    verbs indicating ability or potential outcomes

    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
    201
    +0.17
    1.0%
    85
    +0.16
    0.9%
    25
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.17
    0.06
    25
    +0.16
    0.06
    502
    +0.12
    0.06
    Negative Logits
     Pradesh
    -1.95
     heart
    -1.85
    wise
    -1.80
     glance
    -1.76
    abella
    -1.66
    forth
    -1.65
     simplicity
    -1.63
     hearts
    -1.63
     welcome
    -1.63
     heartbeat
    -1.62
    POSITIVE LOGITS
    ī
    1.95
    ĭ
    1.87
    »
    1.69
    ħ
    1.67
    §
    1.64
    Ħ
    1.60
    Ī
    1.58
    olin
    1.58
    Ļª
    1.53
    icago
    1.52
    Activations Density 0.890%

    No Known Activations