© 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. 2366
    Prev
    Next
    INDEX
    Explanations

    words related to data management and information exchange

    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
    50
    +0.09
    0.3%
    1120
    +0.06
    0.2%
    1363
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1871
    +0.09
    0.02
    1894
    +0.06
    0.03
    845
    +0.06
    0.03
    Negative Logits
                
    -0.77
                    
    -0.77
            
    -0.76
    -0.74
                        
    -0.74
    -0.73
    -0.73
                     
    -0.73
     kronor
    -0.72
    Và
    -0.71
    POSITIVE LOGITS
     affor
    2.79
     maneu
    2.58
     increa
    2.50
     scrat
    2.42
     snoopy
    2.41
     strick
    2.38
     shenan
    2.35
     unden
    2.35
     michelin
    2.34
     impra
    2.33
    Activations Density 0.242%

    No Known Activations