© 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. 1399
    Prev
    Next
    INDEX
    Explanations

    mentions of the action of throwing a ball or hitting a ball in sports contexts

    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
    +0.11
    0.5%
    299
    +0.05
    0.2%
    442
    +0.04
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1343
    +0.11
    0.12
    283
    +0.05
    0.00
    1363
    +0.04
    0.06
    Negative Logits
    <bos>
    -1.59
    /**
    -1.03
     intersper
    -0.97
    <?
    -0.96
    ⓧ
    -0.96
    
    -0.96
    <?
    -0.93
    /*
    -0.89
    /***
    -0.86
     incarcer
    -0.85
    POSITIVE LOGITS
     véhic
    1.14
     soulign
    1.03
     affich
    0.96
     délib
    0.96
     quoique
    0.94
     décid
    0.94
     catég
    0.94
     déliv
    0.91
     karton
    0.89
     renfer
    0.89
    Activations Density 1.204%

    No Known Activations