© 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. 11990
    Prev
    Next
    INDEX
    Explanations

    proper nouns and numerical data related to infrastructure projects

    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
    50
    +0.16
    0.5%
    1577
    +0.13
    0.4%
    1531
    +0.13
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1531
    +0.16
    0.10
    1784
    +0.13
    0.11
    1499
    +0.13
    0.13
    Negative Logits
    <bos>
    -0.95
    WriteBarrier
    -0.60
    сылкі
    -0.56
    IGraphics
    -0.52
     saites
    -0.50
     مرئيه
    -0.50
     CreateTagHelper
    -0.49
    tagHelperRunner
    -0.49
    ksikon
    -0.49
    invokeLater
    -0.49
    POSITIVE LOGITS
     inev
    1.13
     sovere
    1.11
     wherea
    1.09
     Khart
    1.08
     stockholm
    1.08
     Juf
    1.06
     maneu
    1.03
     increa
    1.03
     suscep
    1.02
     squa
    1.02
    Activations Density 4.833%

    No Known Activations