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. 6-RES-JB
    6. 5317
    Prev
    Next
    INDEX
    Explanations

    expressions related to excitement or anticipation

    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.6.hook_resid_post_16384_anthropic_fast_lr
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.21
    0.9%
    1967
    +0.20
    0.9%
    1150
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    588
    +0.21
    0.12
    1314
    +0.20
    0.07
    735
    +0.14
    0.07
    Negative Logits
    <bos>
    -2.50
    Vezi
    -0.86
    Πηγή
    -0.84
    Asimismo
    -0.81
    Educación
    -0.78
    Bibliografía
    -0.78
    Véase
    -0.78
    Provincia
    -0.77
    Varför
    -0.77
    LayoutStyle
    -0.76
    POSITIVE LOGITS
     unlaw
    2.05
     reluct
    1.96
     impra
    1.90
     disagre
    1.88
     disreg
    1.85
     affor
    1.84
     maneu
    1.83
     shenan
    1.81
     volunte
    1.81
     unden
    1.81
    Activations Density 4.475%

    No Known Activations