Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 15913
    Prev
    Next
    INDEX
    Explanations

    details related to specific events or incidents, possibly involving conflict or controversy

    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
    184
    +0.15
    0.5%
    2019
    +0.12
    0.4%
    50
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.15
    0.02
    1288
    +0.12
    0.04
    1450
    +0.11
    0.01
    Negative Logits
    <bos>
    -0.93
    تقاوى
    -0.64
    astéroïdes
    -0.57
     Biôgrafia
    -0.56
    новништво
    -0.55
    arrings
    -0.54
     Administrativna
    -0.54
     zvuky
    -0.52
     otomatig
    -0.52
    erenc
    -0.52
    POSITIVE LOGITS
     Haci
    1.10
     Perci
    1.00
     Khart
    0.92
     Simult
    0.91
     Conclu
    0.91
     inconce
    0.90
    Glej
    0.88
     Juf
    0.87
     reconno
    0.85
     apprehen
    0.85
    Activations Density 0.448%

    No Known Activations