© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 11960
    Prev
    Next
    INDEX
    Explanations

    positive and motivating language, including mentions of positive community building and support

    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
    1438
    +0.13
    0.4%
    1978
    +0.10
    0.3%
    1343
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    282
    +0.13
    0.03
    244
    +0.10
    0.04
    1820
    +0.10
    0.03
    Negative Logits
     depic
    -0.81
     gaily
    -0.76
     disagre
    -0.70
     poetical
    -0.69
     madonna
    -0.67
     increa
    -0.66
     stratigraph
    -0.65
     createdBy
    -0.65
     sophistic
    -0.65
     impra
    -0.65
    POSITIVE LOGITS
    cèse
    0.54
     ролях
    0.53
     restera
    0.53
     contribue
    0.53
     accepte
    0.49
     تجمعات
    0.48
     estudia
    0.47
    انظر
    0.47
    InjectMocks
    0.47
     contribuye
    0.46
    Activations Density 0.331%

    No Known Activations