© 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. 708
    Prev
    Next
    INDEX
    Explanations

    commentary or opinions in a text

    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
    1604
    +0.08
    0.3%
    486
    +0.07
    0.2%
    1870
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1291
    +0.08
    0.02
    1689
    +0.07
    0.02
    1363
    +0.07
    0.02
    Negative Logits
    <bos>
    -1.10
    <?
    -0.77
    glEnd
    -0.74
    EndInit
    -0.73
    terraform
    -0.71
    GIVEREF
    -0.70
    ⓧ
    -0.70
    
    -0.69
    xnn
    -0.69
    execSQL
    -0.69
    POSITIVE LOGITS
     commentary
    2.36
     maneu
    2.31
     increa
    2.20
     commentaries
    2.20
     Commentary
    2.15
     depic
    2.06
     reluct
    2.06
     disagre
    2.03
     stockholm
    2.03
     volunte
    2.01
    Activations Density 0.065%

    No Known Activations