© 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. 599
    Prev
    Next
    INDEX
    Explanations

    words related to a specific brand or person, particularly focusing on repeated mentions of 'Hardy.'

    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
    1276
    +0.10
    0.3%
    1983
    +0.08
    0.2%
    899
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1120
    +0.10
    0.05
    690
    +0.08
    0.05
    1343
    +0.08
    0.04
    Negative Logits
     ми
    -0.66
    ator
    -0.63
    txt
    -0.60
     бар
    -0.60
     ज
    -0.60
     рас
    -0.60
     с
    -0.60
    คร
    -0.60
    ним
    -0.59
     لح
    -0.59
    POSITIVE LOGITS
     Hardy
    2.54
    Hardy
    2.33
     ftu
    1.94
     lidl
    1.91
     guarante
    1.88
     effe
    1.87
     sovere
    1.81
     desir
    1.80
     milano
    1.80
     fep
    1.79
    Activations Density 0.297%

    No Known Activations