© 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. 13306
    Prev
    Next
    INDEX
    Explanations

    conditional phrases indicating observation or explanation

    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
    1510
    +0.09
    0.2%
    1343
    +0.07
    0.2%
    2000
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1801
    +0.09
    0.01
    1136
    +0.07
    0.02
    1551
    +0.07
    0.02
    Negative Logits
     Augu
    -1.02
     EEU
    -0.94
     Juf
    -0.94
     unve
    -0.93
     Simult
    -0.93
     unlaw
    -0.91
     Eft
    -0.88
     increa
    -0.88
     Intere
    -0.88
     depic
    -0.87
    POSITIVE LOGITS
     CURLOPT
    0.60
    InjectAttribute
    0.55
     understand
    0.50
     săptăm
    0.50
    ɵɵ
    0.49
    PDOException
    0.49
    understand
    0.48
    getSystemService
    0.48
    ContentLoaded
    0.48
    CURLOPT
    0.48
    Activations Density 0.103%

    No Known Activations