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. 758
    Prev
    Next
    INDEX
    Explanations

    instances where the word "too" is used

    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
    200
    +0.15
    0.5%
    9
    +0.14
    0.5%
    674
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    200
    +0.15
    0.04
    9
    +0.14
    0.04
    370
    +0.12
    0.04
    Negative Logits
     maksi
    -0.89
     akku
    -0.80
     naer
    -0.76
     seksi
    -0.76
     häm
    -0.72
     osal
    -0.71
     keramik
    -0.71
     mahd
    -0.70
     indik
    -0.69
     kela
    -0.69
    POSITIVE LOGITS
    too
    0.84
     TOO
    0.80
     too
    0.79
     Too
    0.74
    Too
    0.74
    TOO
    0.71
    jednoc
    0.54
     camaro
    0.54
     demasiado
    0.49
     allzu
    0.49
    Activations Density 0.044%

    No Known Activations