© 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. 6-RES-JB
    11. 8497
    Prev
    Next
    INDEX
    Explanations

    architectural firm names and design-related terms

    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.6.hook_resid_post_16384_anthropic_fast_lr
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.24
    0.9%
    1385
    +0.11
    0.4%
    872
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1385
    +0.24
    0.08
    872
    +0.11
    0.06
    882
    +0.09
    0.05
    Negative Logits
    <bos>
    -2.56
    ⓧ
    -0.99
    <?
    -0.76
    /**
    -0.68
    
    -0.65
    💼
    -0.64
    /*++
    -0.60
    Демографія
    -0.60
    /*
    -0.59
    WriteAttribute
    -0.58
    POSITIVE LOGITS
     effe
    1.20
     wien
    1.20
     lidl
    1.17
     increa
    1.17
     inev
    1.17
     Minang
    1.16
     eiffel
    1.14
     maroc
    1.13
     peppa
    1.12
     desir
    1.11
    Activations Density 1.093%

    No Known Activations