© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 399
    Prev
    Next
    INDEX
    Explanations

    descriptions of architectural features and professional titles of individuals designing buildings

    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
    1325
    +0.08
    0.3%
    50
    +0.08
    0.3%
    528
    +0.07
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1385
    +0.08
    0.03
    1798
    +0.08
    0.03
    1870
    +0.07
    0.00
    Negative Logits
    <bos>
    -1.27
     shuddered
    -0.74
    
    -0.73
     reaff
    -0.68
     intensify
    -0.67
     denounce
    -0.66
     deepen
    -0.65
     appear
    -0.65
     shivered
    -0.64
     abolish
    -0.64
    POSITIVE LOGITS
     catég
    1.47
     véhic
    1.44
     Khart
    1.39
     maneu
    1.39
     Juf
    1.38
     quoique
    1.38
     vété
    1.32
     eyel
    1.32
     habile
    1.31
     désol
    1.30
    Activations Density 0.249%

    No Known Activations