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

    Neuronpedia

    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. 13226
    Prev
    Next
    INDEX
    Explanations

    words related to personal qualities and characteristics

    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
    478
    +0.16
    0.5%
    1385
    +0.14
    0.5%
    1741
    +0.14
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    131
    +0.16
    0.05
    332
    +0.14
    0.06
    478
    +0.14
    0.06
    Negative Logits
     depic
    -1.92
     reluct
    -1.85
     maneu
    -1.75
     disagre
    -1.75
     encomp
    -1.73
     gaily
    -1.72
     inev
    -1.72
     erad
    -1.71
     excru
    -1.71
     accla
    -1.71
    POSITIVE LOGITS
     ability
    0.84
     own
    0.82
     abilities
    0.73
     life
    0.68
     overall
    0.67
     capabilities
    0.67
     entire
    0.64
     future
    0.61
     biggest
    0.61
     relationship
    0.61
    Activations Density 0.365%

    No Known Activations