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. 10044
    Prev
    Next
    INDEX
    Explanations

    text related to domestic violence, including descriptions, discussions, and ways to seek help

    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
    604
    +0.15
    0.4%
    674
    +0.12
    0.3%
    509
    +0.11
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    225
    +0.15
    0.05
    509
    +0.12
    0.07
    109
    +0.11
    0.05
    Negative Logits
     disagre
    -1.54
     maneu
    -1.48
     apprehen
    -1.47
     depic
    -1.44
     encomp
    -1.41
     impra
    -1.40
     shenan
    -1.40
     reluct
    -1.39
     uninten
    -1.38
     gaily
    -1.38
    POSITIVE LOGITS
     condolences
    0.68
    AsUp
    0.67
    gdala
    0.67
     GeoNames
    0.66
     tragedy
    0.66
     uska
    0.64
     compassion
    0.63
    Tē
    0.63
    ⇑
    0.62
     Exacts
    0.61
    Activations Density 0.933%

    No Known Activations