© 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. 1301
    Prev
    Next
    INDEX
    Explanations

    titles or headings that are considered to be descriptive of the content they introduce

    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
    528
    +0.08
    0.3%
    900
    +0.08
    0.2%
    977
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    977
    +0.08
    0.03
    1091
    +0.08
    0.02
    109
    +0.06
    0.02
    Negative Logits
    ⓧ
    -0.78
    
    -0.70
    <bos>
    -0.68
    /*
    -0.67
     LoggerFactory
    -0.63
    databind
    -0.63
    <?
    -0.61
    ImageField
    -0.61
    /**
    -0.58
    clear
    -0.57
    POSITIVE LOGITS
     entitled
    2.54
    entitled
    2.29
     entitle
    1.80
     stockholm
    1.69
     entitlement
    1.56
     wien
    1.54
     lidl
    1.50
     franz
    1.49
     budapest
    1.46
     aen
    1.45
    Activations Density 0.159%

    No Known Activations