© 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 EvalsExportsinterp-engineNEWAPI 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. 29436
    Prev
    Next
    INDEX
    Explanations

    instances of the word "chat" and related terms suggesting discussion or conversation

    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
    +0.19
    1.1%
    156
    +0.17
    1.0%
    53
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    341
    +0.19
    0.01
    286
    +0.17
    0.01
    416
    +0.12
    0.01
    Negative Logits
     Antarctic
    -1.67
     glycol
    -1.50
     Hg
    -1.46
     SAS
    -1.43
     ureth
    -1.41
     Antar
    -1.40
     href
    -1.34
     phosphate
    -1.33
     sweat
    -1.30
     paper
    -1.30
    POSITIVE LOGITS
    rooms
    2.01
    oon
    1.93
    ty
    1.79
    ues
    1.78
    room
    1.74
    manship
    1.67
    bare
    1.63
    piece
    1.57
    stro
    1.56
     scenes
    1.50
    Activations Density 0.015%

    No Known Activations