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. 15046
    Prev
    Next
    INDEX
    Explanations

    sentences related to ethical values, social justice, and political commentary

    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
    80
    +0.09
    0.2%
    872
    +0.09
    0.2%
    1135
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    382
    +0.09
    0.05
    163
    +0.09
    0.02
    1135
    +0.08
    0.03
    Negative Logits
     suspic
    -1.50
     excru
    -1.47
     reluct
    -1.43
     embra
    -1.41
     Perci
    -1.40
     inev
    -1.38
     accla
    -1.37
     impra
    -1.37
     compen
    -1.36
     increa
    -1.34
    POSITIVE LOGITS
     ones
    0.86
     which
    0.69
    ones
    0.66
     Ones
    0.63
     whose
    0.63
    which
    0.62
     including
    0.62
     where
    0.61
     sahiptir
    0.60
     olyan
    0.59
    Activations Density 0.314%

    No Known Activations