© 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. 1012
    Prev
    Next
    INDEX
    Explanations

    potential issues or errors and providing solutions or explanations in a response 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
    50
    +0.09
    0.3%
    141
    +0.07
    0.2%
    1589
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    2037
    +0.09
    0.04
    468
    +0.07
    0.03
    1363
    +0.06
    0.04
    Negative Logits
    <bos>
    -0.79
    lccn
    -0.67
    /***
    -0.59
    Cif
    -0.58
     fend
    -0.58
    buenos
    -0.56
    Parte
    -0.56
    Rusia
    -0.53
    glBind
    -0.52
    unitOfWork
    -0.52
    POSITIVE LOGITS
     lele
    1.38
     jaya
    1.29
     aen
    1.28
     jawa
    1.21
     bandung
    1.21
     fua
    1.21
     Minang
    1.14
     dises
    1.11
     Keny
    1.11
     philanth
    1.10
    Activations Density 0.285%

    No Known Activations