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. 15356
    Prev
    Next
    INDEX
    Explanations

    strings of text characters that resemble commands or communication in a digital context

    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
    776
    +0.14
    0.4%
    1343
    +0.12
    0.3%
    453
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1786
    +0.14
    0.03
    776
    +0.12
    0.04
    1297
    +0.10
    0.03
    Negative Logits
    <bos>
    -0.78
    RegressionTest
    -0.71
    OrNil
    -0.71
    Autoritní
    -0.69
    verwijspagina
    -0.65
     ivelany
    -0.65
    Podob
    -0.61
    Prí
    -0.61
    Jako
    -0.61
    Демографія
    -0.60
    POSITIVE LOGITS
     keramik
    1.00
     lancia
    0.96
     maksi
    0.92
     silikon
    0.91
     kafe
    0.91
     melat
    0.89
     seksi
    0.85
     akus
    0.85
     mikrofon
    0.85
     alkoh
    0.84
    Activations Density 0.089%

    No Known Activations