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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 526
    Prev
    Next
    INDEX
    Explanations

    expressions related to gender identity and societal norms

    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
    674
    +0.14
    0.4%
    1334
    +0.13
    0.4%
    1510
    +0.11
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1334
    +0.14
    0.09
    1415
    +0.13
    0.05
    920
    +0.11
    0.05
    Negative Logits
     fei
    -0.97
     Græ
    -0.97
     lele
    -0.94
     plak
    -0.93
     gend
    -0.90
     wien
    -0.90
     kompakt
    -0.90
     mef
    -0.87
     Mâ
    -0.87
     „,
    -0.86
    POSITIVE LOGITS
     avoid
    0.94
     ensure
    0.90
     prevent
    0.85
     protect
    0.81
     conserve
    0.79
     improve
    0.79
     achieve
    0.79
     maximize
    0.77
     compensate
    0.75
     facilitate
    0.74
    Activations Density 0.274%

    No Known Activations