© 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. 1529
    Prev
    Next
    INDEX
    Explanations

    references to specific locations, such as the Capitol Hill

    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
    1757
    +0.12
    0.4%
    314
    +0.08
    0.3%
    1506
    +0.08
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    855
    +0.12
    0.02
    1506
    +0.08
    0.02
    1613
    +0.08
    0.02
    Negative Logits
    /**
    -0.86
    /*
    -0.68
     kiedyś
    -0.64
    <?
    -0.61
    ⓧ
    -0.59
     make
    -0.55
     draw
    -0.53
     подо
    -0.53
     once
    -0.53
    排
    -0.52
    POSITIVE LOGITS
     Capitol
    2.71
    Capitol
    2.39
     capitol
    1.78
     CAPIT
    1.45
     pank
    1.33
     kark
    1.33
     saar
    1.28
     Juf
    1.26
     naer
    1.26
     dovr
    1.24
    Activations Density 0.246%

    No Known Activations