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. 236
    Prev
    Next
    INDEX
    Explanations

    phrases related to experiences or possession of certain qualities

    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
    1096
    +0.15
    0.5%
    1392
    +0.14
    0.5%
    32
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1096
    +0.15
    0.06
    1392
    +0.14
    0.07
    32
    +0.13
    0.06
    Negative Logits
     parlar
    -0.52
    zsef
    -0.50
     Girolamo
    -0.49
     Sén
    -0.49
     ekos
    -0.49
     Ministero
    -0.45
     pluri
    -0.45
     Parco
    -0.45
    Bacterial
    -0.45
    ologio
    -0.44
    POSITIVE LOGITS
    having
    0.83
     HAVING
    0.82
     having
    0.76
     HAVE
    0.69
    have
    0.69
    Having
    0.68
    HAVE
    0.68
     Having
    0.66
    HAV
    0.65
     daudz
    0.59
    Activations Density 0.226%

    No Known Activations