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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 2400
    Prev
    Next
    INDEX
    Explanations

    mentions of educational training courses and forms related to course development

    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
    +0.15
    0.5%
    674
    +0.09
    0.3%
    1654
    +0.08
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1438
    +0.15
    0.05
    908
    +0.09
    0.02
    2044
    +0.08
    0.06
    Negative Logits
     secon
    -1.08
     mef
    -1.05
     squa
    -0.98
     socie
    -0.95
     tramont
    -0.92
     meis
    -0.92
     lyon
    -0.92
     fte
    -0.90
     dises
    -0.89
     unce
    -0.89
    POSITIVE LOGITS
    toPromise
    0.58
    спользова
    0.51
    টি
    0.49
     which
    0.49
    udesta
    0.47
    QUADS
    0.47
     that
    0.45
     constituyen
    0.44
     containing
    0.44
    спользу
    0.43
    Activations Density 0.439%

    No Known Activations