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. 0-RES-JB
    6. 28
    Prev
    Next
    INDEX
    Explanations

    mentions of business cards

    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.0.hook_resid_post_16384_anthropic
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    411
    +0.08
    0.3%
    1909
    +0.08
    0.3%
    690
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1356
    +0.08
    0.04
    158
    +0.08
    0.03
    217
    +0.07
    0.03
    Negative Logits
    <bos>
    -1.54
    protected
    -0.71
    useEffect
    -0.71
    public
    -0.71
    var
    -0.70
    /**
    
    -0.69
    const
    -0.68
    ///
    -0.68
    onClick
    -0.68
    arr
    -0.66
    POSITIVE LOGITS
     maneu
    2.06
     impra
    1.82
     affor
    1.80
     reluct
    1.77
     accla
    1.76
     increa
    1.76
     fortn
    1.73
     unlaw
    1.72
     inev
    1.71
     stockholm
    1.70
    Activations Density 0.114%

    No Known Activations