© 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
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 12524
    Prev
    Next
    INDEX
    Explanations

    references to legislative bills and measures

    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
    1499
    +0.10
    0.3%
    1870
    +0.10
    0.3%
    776
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1499
    +0.10
    0.07
    1224
    +0.10
    0.05
    1973
    +0.10
    0.04
    Negative Logits
     fluo
    -1.48
     doman
    -1.37
     intermitt
    -1.37
     deleter
    -1.37
     nutr
    -1.36
     secon
    -1.35
     thermomix
    -1.34
     trico
    -1.33
     overla
    -1.31
     seiz
    -1.30
    POSITIVE LOGITS
     itself
    0.80
    টি
    0.74
     will
    0.67
    Conclusión
    0.66
     proposed
    0.66
     proposal
    0.66
     seems
    0.64
     includes
    0.63
     allows
    0.62
     does
    0.62
    Activations Density 0.353%

    No Known Activations