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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 533
    Prev
    Next
    INDEX
    Explanations

    text related to medical conditions, especially illnesses and their implications

    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
    1678
    +0.08
    0.3%
    812
    +0.08
    0.3%
    50
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1244
    +0.08
    0.02
    861
    +0.08
    0.02
    468
    +0.07
    0.02
    Negative Logits
    <bos>
    -1.62
    lateinit
    -0.59
    //---
    -0.59
    //{
    -0.56
    public
    -0.54
     continue
    -0.54
    <!--
    -0.54
    Bande
    -0.53
     Gri
    -0.52
    continue
    -0.50
    POSITIVE LOGITS
     illness
    1.81
     Illness
    1.69
     illnesses
    1.48
     Minang
    1.47
     jaya
    1.46
     bandung
    1.38
     sickness
    1.30
     jawa
    1.24
     Bekasi
    1.20
     Karang
    1.16
    Activations Density 0.084%

    No Known Activations