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-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 10156
    Prev
    Next
    INDEX
    Explanations

    sentences that convey strong opinions or conclusions, often expressed with emphasis

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1984
    +0.12
    0.4%
    1491
    +0.10
    0.3%
    11
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    871
    +0.12
    0.06
    1984
    +0.10
    0.07
    11
    +0.10
    0.06
    Negative Logits
    Må
    -0.57
    Conheça
    -0.55
    Apre
    -0.53
     turi
    -0.52
    Zdra
    -0.52
    Vía
    -0.51
    Meto
    -0.51
    Pró
    -0.50
    Espa
    -0.50
    Conoce
    -0.50
    POSITIVE LOGITS
     unden
    1.09
     suscep
    1.07
     purcha
    1.05
     lorenzo
    1.05
    <bos>
    1.04
     nicolas
    1.04
     jacques
    1.02
     jorge
    0.99
     increa
    0.96
     tetrach
    0.96
    Activations Density 0.412%

    No Known Activations