© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 1974
    Prev
    Next
    INDEX
    Explanations

    words related to official organizations or government agencies, particularly including "Bureau."

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1407
    +0.11
    0.4%
    1937
    +0.10
    0.3%
    687
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    800
    +0.11
    0.02
    337
    +0.10
    0.02
    1685
    +0.08
    0.02
    Negative Logits
    <bos>
    -0.69
     unload
    -0.53
     disarm
    -0.53
    
    
    -0.52
    //---
    -0.52
    /**
    -0.51
    }(),
    -0.50
    énégal
    -0.50
    /*++
    -0.50
     invade
    -0.49
    POSITIVE LOGITS
     Bureau
    2.92
    Bureau
    2.63
     bureau
    2.55
     BUREAU
    2.34
     bureaus
    1.89
    bureau
    1.84
     Bure
    1.50
    UREAU
    1.33
     bureaux
    1.28
     bure
    0.94
    Activations Density 0.187%

    No Known Activations