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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 12964
    Prev
    Next
    INDEX
    Explanations

    references to specific groups such as Muslims, immigrants, Americans, GMO advocates, Russians, Assad supporters, and Catholics in a negative context

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1870
    +0.11
    0.3%
    2034
    +0.11
    0.3%
    1741
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    74
    +0.11
    0.03
    1755
    +0.11
    0.03
    1539
    +0.10
    0.03
    Negative Logits
     sputnik
    -0.85
     tricot
    -0.84
     unden
    -0.84
     pessi
    -0.82
     swarovski
    -0.81
     disgra
    -0.81
     inappro
    -0.79
     exorbit
    -0.79
     tupperware
    -0.78
     madonna
    -0.78
    POSITIVE LOGITS
    <bos>
    0.53
     sentiment
    0.51
     sentiments
    0.51
     anti
    0.49
    колеп
    0.48
    IntoConstraints
    0.47
     advocate
    0.46
     policies
    0.46
    नलिखित
    0.45
     campaign
    0.45
    Activations Density 0.131%

    No Known Activations