© 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. 10-RES-JB
    11. 617
    Prev
    Next
    INDEX
    Explanations

    references to military aid and defense-related topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.10.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    chanind/openwebtext-gemma
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    1,024
    Dataset
    ctigges/openwebtext-gemma-1024-cl
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.31
    1.1%
    1177
    +0.10
    0.4%
    604
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1048
    +0.31
    0.06
    939
    +0.10
    0.08
    1499
    +0.09
    0.07
    Negative Logits
    <bos>
    -2.87
     intersper
    -0.93
     amass
    -0.72
     mena
    -0.71
     pessi
    -0.69
     underval
    -0.69
     unify
    -0.68
    /***
    -0.66
     boh
    -0.66
     crou
    -0.65
    POSITIVE LOGITS
     conçue
    1.02
     élégante
    0.85
     destinées
    0.82
     conçus
    0.82
     équipée
    0.82
     soulign
    0.81
     conç
    0.78
     tucson
    0.78
     Estaba
    0.77
    atguigu
    0.75
    Activations Density 0.969%

    No Known Activations