© 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
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 2029
    Prev
    Next
    INDEX
    Explanations

    text related to poster presentations at conferences or contests

    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
    1671
    +0.09
    0.3%
    645
    +0.08
    0.3%
    1865
    +0.08
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1865
    +0.09
    0.03
    645
    +0.08
    0.03
    214
    +0.08
    0.02
    Negative Logits
    <?
    -0.86
    <bos>
    -0.85
    ⓧ
    -0.76
    /***
    -0.73
    
    -0.67
    /**
    -0.64
    ///**
    -0.64
    //---
    -0.58
    /*
    -0.57
    KEYCODE
    -0.56
    POSITIVE LOGITS
     poster
    2.63
     posters
    2.36
     Poster
    2.30
    poster
    2.20
    Poster
    2.16
     Posters
    2.10
    Posters
    1.88
    posters
    1.82
     hcm
    1.23
     Minang
    1.10
    Activations Density 0.095%

    No Known Activations