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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27938
    Prev
    Next
    INDEX
    Explanations

    responses or replies within a technical or formal context

    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
    +0.20
    1.2%
    376
    +0.15
    0.8%
    413
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    320
    +0.20
    0.01
    413
    +0.15
    0.01
    126
    +0.12
    0.01
    Negative Logits
    ĥ½
    -2.23
    roe
    -1.88
    itoring
    -1.67
    usterity
    -1.56
     its
    -1.54
    arrant
    -1.54
    ¿½
    -1.53
    |$.
    -1.53
    oxacin
    -1.50
     UAE
    -1.48
    POSITIVE LOGITS
    nier
    1.68
    able
    1.55
    fix
    1.54
    starter
    1.50
    ató
    1.47
     Manager
    1.46
     includes
    1.42
    ably
    1.40
    GIS
    1.36
    ily
    1.35
    Activations Density 0.019%

    No Known Activations