© 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. 14248
    Prev
    Next
    INDEX
    Explanations

    phrases indicating trial and error or experimentation in a technical or scientific context, particularly related to a specific procedure or process

    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
    604
    +0.09
    0.3%
    764
    +0.08
    0.2%
    626
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1919
    +0.09
    0.06
    1415
    +0.08
    0.03
    1513
    +0.07
    0.01
    Negative Logits
     fta
    -1.18
     fte
    -1.17
     guarante
    -1.10
     oun
    -1.09
     NOO
    -1.09
     thut
    -1.08
     aen
    -1.06
     ftu
    -1.06
     soigne
    -1.06
     ?...
    -1.05
    POSITIVE LOGITS
     prefer
    0.65
     decided
    0.64
     opted
    0.62
     discontinued
    0.61
     instead
    0.60
     discontinue
    0.59
     inkább
    0.58
     withdrew
    0.58
     opt
    0.58
     refrain
    0.57
    Activations Density 0.627%

    No Known Activations