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

    Neuronpedia

    Jacobian LensNEW
    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. 10-RES-JB
    11. 2552
    Prev
    Next
    INDEX
    Explanations

    references to doctrines or beliefs in various contexts

    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
    1937
    +0.17
    1.0%
    1034
    +0.16
    0.9%
    1047
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1363
    +0.17
    0.02
    1137
    +0.16
    0.02
    144
    +0.13
    0.01
    Negative Logits
    <bos>
    -2.08
     كومونز
    -0.74
     himo
    -0.71
    be
    -0.63
    mi
    -0.63
    min
    -0.62
    av
    -0.61
    /**
    -0.61
                            
    -0.61
    AnchorStyles
    -0.59
    POSITIVE LOGITS
     impra
    1.61
     increa
    1.58
     Juf
    1.54
     affor
    1.54
     reluct
    1.50
     disagre
    1.49
     stockholm
    1.48
     maneu
    1.48
     unwarran
    1.47
     toledo
    1.46
    Activations Density 0.073%

    No Known Activations