© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 12987
    Prev
    Next
    INDEX
    Explanations

    references to specific names or titles related to popular book and movie series, particularly "Harry Potter."

    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
    479
    +0.17
    0.7%
    204
    +0.16
    0.7%
    1491
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    204
    +0.17
    0.02
    479
    +0.16
    0.02
    1335
    +0.13
    0.03
    Negative Logits
     للاسماء
    -0.67
    mistak
    -0.61
     lenker
    -0.56
    Może
    -0.55
    Dziękuję
    -0.52
     oreo
    -0.51
    Pozdrawiam
    -0.50
    municipi
    -0.50
     swarovski
    -0.49
    zupeł
    -0.49
    POSITIVE LOGITS
     Harry
    1.38
    Harry
    1.33
    HARRY
    1.29
     HARRY
    1.21
    harry
    1.19
     harry
    1.16
     Potter
    0.97
    Potter
    0.91
     wizard
    0.88
     Wizard
    0.84
    Activations Density 0.140%

    No Known Activations