© 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. Residuals
    9. 1-RES-SM
    11. 28641
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Public" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    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
    101
    +0.13
    0.7%
    503
    +0.12
    0.7%
    71
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    101
    +0.13
    0.02
    503
    +0.12
    0.02
    435
    +0.10
    0.02
    Negative Logits
     sexes
    -1.74
     strongest
    -1.44
     predictive
    -1.42
    omers
    -1.41
    rian
    -1.39
    opic
    -1.37
     her
    -1.37
     traditional
    -1.36
     brightest
    -1.36
    rapeut
    -1.33
    POSITIVE LOGITS
    ourse
    1.65
    fiddle
    1.59
    ités
    1.59
    ibus
    1.59
    IFIC
    1.56
    LEY
    1.55
    license
    1.55
    ité
    1.54
     License
    1.53
    mouth
    1.49
    Activations Density 0.057%

    No Known Activations