© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 26964
    Prev
    Next
    INDEX
    Explanations

    the word "villa" with high emphasis, suggesting a focus on specific types of locations

    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
    0.9%
    385
    +0.14
    0.8%
    286
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    385
    +0.15
    0.02
    205
    +0.14
    0.02
    109
    +0.14
    0.02
    Negative Logits
     ?"
    -2.15
    right
    -1.78
     worse
    -1.74
     why
    -1.68
     anyone
    -1.60
    rceil
    -1.59
     anybody
    -1.57
     wrong
    -1.53
    "?"
    -1.49
    soever
    -1.42
    POSITIVE LOGITS
    ue
    1.75
    igent
    1.72
    vue
    1.64
    enos
    1.62
    enstein
    1.58
    âĢij
    1.56
    eries
    1.50
    uncher
    1.47
    election
    1.47
    TeX
    1.46
    Activations Density 1.647%

    No Known Activations