© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 28782
    Prev
    Next
    INDEX
    Explanations

    instances of the word "included."

    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
    376
    +0.20
    1.1%
    463
    +0.13
    0.7%
    14
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    216
    +0.20
    0.03
    505
    +0.13
    0.03
    372
    +0.12
    0.03
    Negative Logits
     celebrated
    -1.78
    »¿
    -1.60
    chron
    -1.44
    §
    -1.44
    grade
    -1.40
    craft
    -1.39
     TODAY
    -1.38
     tomorrow
    -1.36
    earth
    -1.34
     share
    -1.33
    POSITIVE LOGITS
    aking
    1.66
     cartridges
    1.51
    za
    1.49
     pores
    1.42
    anting
    1.38
    zc
    1.38
     defeats
    1.37
    ays
    1.36
     pockets
    1.32
    ational
    1.31
    Activations Density 2.878%

    No Known Activations