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. 15267
    Prev
    Next
    INDEX
    Explanations

    verbs related to rejection or defiance

    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
    1145
    +0.10
    0.3%
    1472
    +0.09
    0.3%
    1065
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1472
    +0.10
    0.02
    1171
    +0.09
    0.03
    1145
    +0.09
    0.03
    Negative Logits
     utop
    -0.54
     solidar
    -0.53
    Mä
    -0.52
     Veter
    -0.51
     trasforma
    -0.51
     ferait
    -0.50
     ouvre
    -0.50
     patrio
    -0.49
     Gep
    -0.49
     amat
    -0.48
    POSITIVE LOGITS
     Refuse
    0.90
     refuse
    0.83
     refusal
    0.75
     refusing
    0.74
     refuses
    0.74
     refused
    0.72
     vainly
    0.71
     gaily
    0.70
     unspeak
    0.65
    Wię
    0.64
    Activations Density 0.071%

    No Known Activations