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 SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 12-RES-JB
    6. 7767
    Prev
    Next
    INDEX
    Explanations

    phrases related to educational assessments and student confidence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.12.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    12
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.37
    1.3%
    738
    +0.10
    0.3%
    509
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1551
    +0.37
    0.05
    1567
    +0.10
    0.04
    212
    +0.10
    0.03
    Negative Logits
    <bos>
    -1.53
    addPreferredGap
    -0.53
    insertBefore
    -0.53
    Enllaços
    -0.50
    addGap
    -0.49
    config
    -0.49
    Ответ
    -0.49
     eius
    -0.49
    $_['
    -0.47
    ByVersion
    -0.47
    POSITIVE LOGITS
     maneu
    1.51
     Minang
    1.50
     disreg
    1.44
     bandung
    1.35
     shenan
    1.35
     chrysler
    1.35
     affor
    1.33
     Jambi
    1.32
     impra
    1.31
     reluct
    1.31
    Activations Density 1.263%

    No Known Activations