© 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. 29321
    Prev
    Next
    INDEX
    Explanations

    the word "another" and its context in various sentences

    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
    402
    +0.16
    0.9%
    225
    +0.13
    0.7%
    315
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    248
    +0.16
    0.08
    278
    +0.13
    0.10
    421
    +0.12
    0.07
    Negative Logits
     Office
    -1.59
    ards
    -1.50
    bent
    -1.49
    supplementary
    -1.48
    implementation
    -1.46
     efforts
    -1.46
    Conclusion
    -1.45
     discussions
    -1.40
    hele
    -1.39
    Conclusions
    -1.38
    POSITIVE LOGITS
    kins
    1.57
     que
    1.45
    mia
    1.42
     huh
    1.38
    ylvania
    1.38
     Rouge
    1.35
    acing
    1.34
     )"
    1.34
    kin
    1.33
     infiltr
    1.32
    Activations Density 0.436%

    No Known Activations