© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 2821
    Prev
    Next
    INDEX
    Explanations

    dome-shaped structures or objects

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    1964
    +0.12
    0.4%
    703
    +0.07
    0.2%
    1065
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1343
    +0.12
    0.03
    1120
    +0.07
    0.02
    1385
    +0.07
    0.02
    Negative Logits
    <bos>
    -0.78
    jetbrains
    -0.55
    glBegin
    -0.54
    FormTagHelper
    -0.54
    β
    -0.51
    cvtColor
    -0.51
     Oatmeal
    -0.51
    annotation
    -0.51
    ույ
    -0.50
    เช
    -0.49
    POSITIVE LOGITS
     dome
    2.70
     Dome
    2.63
     domes
    2.33
    Dome
    2.29
    dome
    2.17
     Domes
    1.69
     Juf
    1.16
     bangkok
    1.14
     cúp
    1.12
     pixabay
    1.10
    Activations Density 0.197%

    No Known Activations