© 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
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 4014
    Prev
    Next
    INDEX
    Explanations

    instances of male names

    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
    1013
    +0.08
    0.2%
    857
    +0.08
    0.2%
    1
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1887
    +0.08
    0.05
    1
    +0.08
    0.04
    1619
    +0.08
    0.05
    Negative Logits
     strick
    -1.68
     increa
    -1.67
     alre
    -1.66
     ?...
    -1.64
     encomp
    -1.64
     secon
    -1.64
     perfet
    -1.60
     guarante
    -1.60
     squa
    -1.60
     depic
    -1.60
    POSITIVE LOGITS
     performing
    0.77
     doing
    0.77
     perform
    0.76
     sitting
    0.72
     playing
    0.72
     enjoying
    0.71
     standing
    0.69
     coming
    0.68
     wearing
    0.68
     appearing
    0.67
    Activations Density 0.488%

    No Known Activations