© 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. 1005
    Prev
    Next
    INDEX
    Explanations

    religious references and beliefs, particularly related to the issues of moral and political principles, as well as concerns related to Israel

    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
    872
    +0.12
    0.3%
    1042
    +0.08
    0.2%
    1533
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    872
    +0.12
    0.06
    2030
    +0.08
    0.02
    1700
    +0.08
    0.03
    Negative Logits
     intersper
    -0.87
     territo
    -0.86
     dises
    -0.86
     tew
    -0.86
     convenable
    -0.83
     §.
    -0.82
     osal
    -0.82
     thut
    -0.81
     fta
    -0.81
    :,,
    -0.79
    POSITIVE LOGITS
     everything
    0.70
     dominant
    0.70
     priority
    0.69
     sole
    0.69
    everything
    0.64
     dominate
    0.64
    Everything
    0.63
     Everything
    0.63
     alone
    0.60
    priority
    0.59
    Activations Density 0.764%

    No Known Activations