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. 15740
    Prev
    Next
    INDEX
    Explanations

    news articles or blog posts

    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
    227
    +0.12
    0.4%
    872
    +0.12
    0.4%
    1150
    +0.11
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1150
    +0.12
    0.04
    924
    +0.12
    0.07
    1295
    +0.11
    0.06
    Negative Logits
     dè
    -0.81
     meras
    -0.71
     bunda
    -0.67
     gela
    -0.66
     benzin
    -0.63
     salg
    -0.63
     alfab
    -0.62
     nein
    -0.62
    Sqft
    -0.62
    ifrance
    -0.61
    POSITIVE LOGITS
     reluct
    0.93
     disreg
    0.89
     encomp
    0.86
     maneu
    0.85
     accla
    0.84
     disagre
    0.81
     resear
    0.81
     upvoted
    0.80
     uninten
    0.80
     philanth
    0.77
    Activations Density 0.775%

    No Known Activations