© 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. 533
    Prev
    Next
    INDEX
    Explanations

    sentences expressing awareness or reflection on personal circumstances or societal perceptions

    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
    +0.29
    1.1%
    478
    +0.13
    0.5%
    604
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    478
    +0.29
    0.05
    1352
    +0.13
    0.05
    258
    +0.10
    0.05
    Negative Logits
    <bos>
    -2.56
     intersper
    -1.18
     shenan
    -1.02
     endow
    -0.93
     vainly
    -0.93
     encomp
    -0.90
     disambigu
    -0.89
     apprehen
    -0.89
     harmonize
    -0.88
     reconno
    -0.84
    POSITIVE LOGITS
     soggior
    0.79
     paradiso
    0.77
     sposa
    0.77
     predecess
    0.77
     tramont
    0.76
     palio
    0.74
     ridu
    0.74
     marea
    0.73
     eccell
    0.73
     dott
    0.72
    Activations Density 1.559%

    No Known Activations