Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27254
    Prev
    Next
    INDEX
    Explanations

    emotional states, specifically feelings of distress or discomfort

    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
    +0.15
    0.8%
    274
    +0.12
    0.7%
    228
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    396
    +0.15
    0.02
    228
    +0.12
    0.02
    432
    +0.11
    0.02
    Negative Logits
    chez
    -1.60
    plt
    -1.51
    ]>
    -1.51
    TRODUCTION
    -1.50
    TO
    -1.43
    etc
    -1.43
     fame
    -1.42
    cci
    -1.40
    ahi
    -1.39
    iless
    -1.39
    POSITIVE LOGITS
     deficit
    1.73
     deficits
    1.68
     differences
    1.64
    bold
    1.63
     oriented
    1.47
    ahead
    1.39
     Leader
    1.39
    ignment
    1.39
    weight
    1.38
     oper
    1.37
    Activations Density 0.051%

    No Known Activations