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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 10910
    Prev
    Next
    INDEX
    Explanations

    comparative and superlative adjectives related to negative aspects or criticisms

    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
    468
    +0.13
    0.5%
    197
    +0.12
    0.4%
    757
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    197
    +0.13
    0.03
    468
    +0.12
    0.03
    257
    +0.10
    0.03
    Negative Logits
     Veen
    -0.52
     Bohn
    -0.46
     Pott
    -0.46
     Schiller
    -0.44
     Annan
    -0.43
     Gies
    -0.42
     Schot
    -0.41
     Fulda
    -0.41
     Schult
    -0.41
     Eder
    -0.40
    POSITIVE LOGITS
     worst
    1.07
    worst
    1.03
    Worst
    1.00
     worse
    1.00
     Worst
    0.99
    worse
    0.97
     Worse
    0.89
    Worse
    0.87
     jorge
    0.69
     worsened
    0.68
    Activations Density 0.082%

    No Known Activations