© 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. 3912
    Prev
    Next
    INDEX
    Explanations

    descriptive terms related to negative or troublesome situations

    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
    2034
    +0.13
    0.4%
    394
    +0.11
    0.3%
    630
    +0.11
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.13
    0.10
    2044
    +0.11
    0.09
    630
    +0.11
    0.06
    Negative Logits
     succede
    -0.85
    <bos>
    -0.84
     dimenti
    -0.84
     trovo
    -0.83
     appunto
    -0.81
     rispond
    -0.80
     vorrei
    -0.78
     anzi
    -0.76
     vuol
    -0.76
     voleva
    -0.74
    POSITIVE LOGITS
     because
    0.73
     NDEBUG
    0.62
     since
    0.61
     göm
    0.58
     but
    0.58
    político
    0.58
     Şi
    0.57
     whereby
    0.57
    because
    0.57
     however
    0.57
    Activations Density 0.797%

    No Known Activations