© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 728
    Prev
    Next
    INDEX
    Explanations

    technical terms related to tools or products

    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
    +0.19
    1.7%
    1978
    +0.12
    1.1%
    382
    +0.11
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1978
    +0.19
    0.22
    382
    +0.12
    0.18
    478
    +0.11
    -0.00
    Negative Logits
    <bos>
    -2.18
    ⓧ
    -1.21
    <?
    -1.19
    /**
    -1.16
    
    -1.14
    /*
    -1.01
    <?
    -0.85
     springfox
    -0.84
    /***
    -0.80
    /*++
    -0.75
    POSITIVE LOGITS
     affor
    1.50
     maneu
    1.46
     increa
    1.32
     véhic
    1.29
     stockholm
    1.26
     impra
    1.26
     embodi
    1.25
     lidl
    1.23
     maroc
    1.19
     habile
    1.18
    Activations Density 1.079%

    No Known Activations