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. 16044
    Prev
    Next
    INDEX
    Explanations

    comparative phrases and political terms

    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
    980
    +0.07
    0.2%
    1861
    +0.07
    0.2%
    648
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    441
    +0.07
    0.03
    1081
    +0.07
    0.04
    1934
    +0.07
    0.03
    Negative Logits
     unspeak
    -0.77
     McLaugh
    -0.74
     pamph
    -0.69
     philanth
    -0.69
     sophistic
    -0.66
     McInt
    -0.63
     Shakspeare
    -0.63
     nutella
    -0.62
     disagre
    -0.62
     Reminis
    -0.61
    POSITIVE LOGITS
     individual
    1.43
    individual
    1.27
     individually
    1.21
    Individual
    1.18
     Individual
    1.10
     individuales
    1.04
     individ
    0.99
     INDIVIDUAL
    0.98
    individ
    0.91
     individualized
    0.87
    Activations Density 0.466%

    No Known Activations