© 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 SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 767
    Prev
    Next
    INDEX
    Explanations

    percentage terms in a document

    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
    50
    +0.15
    0.6%
    1891
    +0.05
    0.2%
    838
    +0.04
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1596
    +0.15
    0.04
    1059
    +0.05
    0.04
    21
    +0.04
    0.04
    Negative Logits
    <bos>
    -1.69
    ⓧ
    -0.91
    <?
    -0.86
    
    
    -0.82
    /**
    -0.81
    /*
    -0.80
    <?
    
    -0.68
    //---
    -0.64
    },{
    
    -0.62
     prepare
    -0.62
    POSITIVE LOGITS
     embodi
    1.41
     swarovski
    1.34
     véhic
    1.32
     maneu
    1.28
     délib
    1.27
     affor
    1.27
     scrat
    1.23
     pleins
    1.23
     Juf
    1.20
     ecru
    1.18
    Activations Density 0.086%

    No Known Activations