© 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. 30914
    Prev
    Next
    INDEX
    Explanations

    references to annual or yearly occurrences and statistics

    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
    156
    +0.26
    1.5%
    376
    +0.14
    0.8%
    94
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    94
    +0.26
    0.01
    300
    +0.14
    0.01
    259
    +0.13
    0.01
    Negative Logits
    -2.02
    <|outofrange|>
    -2.02
     ↵
    -2.02
    ↵Č
    -2.02
    -2.02
    -2.02
    ↵
    -2.02
    č↵
    -2.02
     ↵
    -2.02
           ↵
    -2.02
    POSITIVE LOGITS
    "}](#
    1.79
    ophore
    1.42
    ?"
    1.40
    ently
    1.37
    gent
    1.36
     arose
    1.35
     evolved
    1.33
     uttered
    1.30
     drew
    1.29
     manslaughter
    1.29
    Activations Density 0.010%

    No Known Activations