© 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-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 6745
    Prev
    Next
    INDEX
    Explanations

    mentions of legal convictions and criminal activities in news articles

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    860
    +0.13
    0.4%
    1363
    +0.10
    0.4%
    1622
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    860
    +0.13
    0.03
    1363
    +0.10
    0.02
    1622
    +0.10
    0.02
    Negative Logits
     unlaw
    -0.84
     socie
    -0.81
     thermomix
    -0.75
     utop
    -0.75
     desir
    -0.74
     abbiano
    -0.73
     antem
    -0.71
     revan
    -0.70
     dovrebbero
    -0.70
     effe
    -0.69
    POSITIVE LOGITS
     convicted
    1.10
     conviction
    1.04
     convictions
    0.99
     Conv
    0.83
    conv
    0.81
    Conv
    0.79
     convict
    0.72
     convicts
    0.70
     convic
    0.66
     guilty
    0.64
    Activations Density 0.064%

    No Known Activations