© 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. 7645
    Prev
    Next
    INDEX
    Explanations

    mentions of specific racial or ethnic demographics, especially African Americans

    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
    521
    +0.12
    0.3%
    198
    +0.11
    0.3%
    1101
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    521
    +0.12
    0.05
    198
    +0.11
    0.05
    1101
    +0.10
    0.03
    Negative Logits
    ENODEV
    -0.53
    Šaltiniai
    -0.51
    ‿
    -0.48
     besa
    -0.47
    庸
    -0.47
    zczegól
    -0.46
     gynhyrchwyd
    -0.46
    paddingVertical
    -0.46
    ofType
    -0.46
    Aplicaciones
    -0.45
    POSITIVE LOGITS
     Whence
    0.94
     indescri
    0.87
     milf
    0.87
     unspeak
    0.87
     emphat
    0.87
     affor
    0.86
     guarante
    0.85
     yoda
    0.83
     nigger
    0.83
     maneu
    0.83
    Activations Density 0.273%

    No Known Activations