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 SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 6-RES-JB
    6. 14853
    Prev
    Next
    INDEX
    Explanations

    legal and regulatory terms and phrases, such as specific laws, acts, regulations, and legal proceedings

    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
    +0.25
    1.0%
    1499
    +0.12
    0.5%
    507
    +0.06
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1499
    +0.25
    0.13
    1941
    +0.12
    0.06
    127
    +0.06
    0.06
    Negative Logits
    <bos>
    -2.93
    ⓧ
    -0.96
    
    
    -0.87
    <?
    -0.83
    intios
    -0.76
    /**
    -0.74
    <?
    
    -0.72
    REASONS
    -0.70
    fflush
    -0.70
    lateinit
    -0.69
    POSITIVE LOGITS
     maneu
    1.83
     Juf
    1.78
     Minang
    1.71
     jaya
    1.69
     Keny
    1.68
     disagre
    1.63
     bandung
    1.63
     affor
    1.63
     accla
    1.60
     hcm
    1.60
    Activations Density 1.408%

    No Known Activations