© 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. 11935
    Prev
    Next
    INDEX
    Explanations

    phrases related to dramatic personal struggles and experiences

    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
    605
    +0.11
    0.4%
    1133
    +0.10
    0.3%
    569
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    569
    +0.11
    0.02
    1133
    +0.10
    0.02
    220
    +0.09
    0.02
    Negative Logits
     §.
    -0.72
     Kategor
    -0.72
     antik
    -0.71
     kompati
    -0.70
     induk
    -0.69
     Præ
    -0.69
     disting
    -0.65
     vernac
    -0.64
     kosme
    -0.64
     Sympathi
    -0.63
    POSITIVE LOGITS
     literally
    1.01
     literal
    0.91
    literally
    0.89
     Literally
    0.85
     literalmente
    0.81
    Literally
    0.80
    solicited
    0.73
    literal
    0.70
    harmed
    0.64
    compromising
    0.59
    Activations Density 0.053%

    No Known Activations