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. 8411
    Prev
    Next
    INDEX
    Explanations

    details about a specific product, focusing on its features and materials

    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
    1385
    +0.19
    0.6%
    1343
    +0.18
    0.6%
    678
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.19
    0.06
    1385
    +0.18
    0.05
    74
    +0.15
    0.02
    Negative Logits
     kemer
    -0.61
    <bos>
    -0.60
     Rasulullah
    -0.59
    AllowAnonymous
    -0.57
    AfterClass
    -0.56
    ServletRequest
    -0.56
    ClientRect
    -0.55
     kontinu
    -0.54
    FormTagHelper
    -0.54
    UnusedPrivate
    -0.54
    POSITIVE LOGITS
     accompagne
    0.81
     soutient
    0.79
     affez
    0.76
     mène
    0.75
     remonte
    0.75
     dirait
    0.74
     trouvera
    0.74
     reconnaît
    0.72
     croit
    0.72
     pymysql
    0.68
    Activations Density 0.247%

    No Known Activations