© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 1363
    Prev
    Next
    INDEX
    Explanations

    phrases expressing strong emphasis or approval

    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
    840
    +0.07
    0.2%
    900
    +0.07
    0.2%
    50
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    840
    +0.07
    0.03
    294
    +0.07
    0.02
    452
    +0.06
    0.02
    Negative Logits
    <bos>
    -1.44
    public
    -0.73
    lateinit
    -0.73
    protected
    -0.67
    <tr>
    -0.67
            
    -0.66
    },{
    -0.65
    <!--
    -0.65
     further
    -0.65
    add
    -0.65
    POSITIVE LOGITS
     darn
    2.52
     damn
    2.27
     DAMN
    2.27
     damned
    2.25
     goddamn
    2.06
     shenan
    2.01
     increa
    1.94
     maneu
    1.92
     excru
    1.91
     freakin
    1.90
    Activations Density 0.087%

    No Known Activations