Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 19168
    Prev
    Next
    INDEX
    Explanations

    details about incidents and disasters

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    481
    +0.20
    0.9%
    714
    +0.13
    0.6%
    720
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    714
    +0.20
    0.05
    384
    +0.13
    0.04
    393
    +0.10
    0.05
    Negative Logits
    essor
    -0.81
    baum
    -0.81
    arten
    -0.80
    oire
    -0.79
    atech
    -0.79
    onomy
    -0.78
    ith
    -0.77
    itsch
    -0.77
    center
    -0.77
    uran
    -0.76
    POSITIVE LOGITS
     starvation
    1.24
     setbacks
    1.23
     deterioration
    1.20
     arrests
    1.19
     shortages
    1.19
     suicides
    1.18
     delays
    1.15
     deaths
    1.15
     layoffs
    1.13
     lawsuits
    1.13
    Activations Density 0.249%

    No Known Activations