© 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 EvalsExportsAPI 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. 14714
    Prev
    Next
    INDEX
    Explanations

    phrases related to speaking up or speaking out on various issues or behalf of others

    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
    791
    +0.12
    0.3%
    47
    +0.10
    0.3%
    1604
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    791
    +0.12
    0.05
    47
    +0.10
    0.04
    538
    +0.09
    0.04
    Negative Logits
     depic
    -0.95
     guarante
    -0.90
     ?...
    -0.86
     accla
    -0.86
     encomp
    -0.85
     increa
    -0.83
     desir
    -0.80
     fta
    -0.80
     fuf
    -0.80
     »>
    -0.79
    POSITIVE LOGITS
     speak
    0.74
     loud
    0.70
     louder
    0.70
     spoken
    0.69
     voice
    0.67
     mouth
    0.63
     speaking
    0.62
     voices
    0.61
    spoken
    0.61
     aloud
    0.60
    Activations Density 0.446%

    No Known Activations