© 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. 1956
    Prev
    Next
    INDEX
    Explanations

    references to items or concepts related to low-quality or unhealthy food

    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
    687
    +0.09
    0.3%
    410
    +0.08
    0.3%
    1637
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1363
    +0.09
    0.03
    1053
    +0.08
    0.02
    452
    +0.07
    0.02
    Negative Logits
    <bos>
    -0.80
    param
    -0.56
     насељу
    -0.56
    <?
    -0.56
    /*
    -0.54
    xnn
    -0.54
     thở
    -0.54
    
    -0.53
    లాలు
    -0.52
    InputBorder
    -0.52
    POSITIVE LOGITS
     junk
    1.36
     Junk
    1.24
     maneu
    1.21
    Junk
    1.13
     increa
    1.08
    junk
    1.05
     gaily
    1.03
     brilli
    1.03
     disreg
    1.00
     impra
    0.99
    Activations Density 0.189%

    No Known Activations