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. 12-RES-JB
    6. 7855
    Prev
    Next
    INDEX
    Explanations

    mentions of the name "Graham."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.12.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    12
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1276
    +0.14
    0.8%
    50
    +0.13
    0.7%
    61
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1741
    +0.14
    0.01
    227
    +0.13
    0.03
    1276
    +0.13
    0.02
    Negative Logits
    <bos>
    -1.33
    awtextra
    -0.62
    dibles
    -0.58
    AutoScaleMode
    -0.56
    ecap
    -0.54
     Hodg
    -0.53
     McLaugh
    -0.53
     Dooley
    -0.52
     Hirst
    -0.51
     Ellison
    -0.50
    POSITIVE LOGITS
     Graham
    1.57
    Graham
    1.50
    graham
    1.21
     graham
    1.19
     GRAHAM
    1.14
     Settembre
    0.85
     tramont
    0.75
     Áng
    0.74
     santiago
    0.72
     palermo
    0.70
    Activations Density 0.197%

    No Known Activations