Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 6-RES-JB
    6. 8853
    Prev
    Next
    INDEX
    Explanations

    personal pronouns and verbs indicating actions or experiences

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.28
    1.4%
    478
    +0.11
    0.6%
    2019
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1510
    +0.28
    0.12
    805
    +0.11
    0.14
    331
    +0.11
    0.13
    Negative Logits
    <bos>
    -3.11
    ⓧ
    -0.71
     springfox
    -0.70
     елның
    -0.70
    RegressionTest
    -0.69
    tonode
    -0.68
    RegistryLite
    -0.67
    EndProject
    -0.66
    
    
    -0.65
     autorytatywna
    -0.65
    POSITIVE LOGITS
     maneu
    0.97
     véhic
    0.96
     soulign
    0.95
    mainwindow
    0.89
     exceptionnelle
    0.88
     surpren
    0.87
     épu
    0.85
     idéale
    0.85
    RetVal
    0.85
     spécialisé
    0.85
    Activations Density 1.615%

    No Known Activations