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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 444
    Prev
    Next
    INDEX
    Explanations

    terms related to locations and reports on social issues, particularly about institutional failures

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    376
    +0.12
    0.5%
    728
    +0.11
    0.5%
    127
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    376
    +0.12
    0.08
    284
    +0.11
    0.08
    728
    +0.11
    0.05
    Negative Logits
    ople
    -0.84
    lished
    -0.83
    ylon
    -0.83
    clips
    -0.81
    phrine
    -0.76
    grading
    -0.73
    dump
    -0.72
    dating
    -0.71
    amphetamine
    -0.71
    tics
    -0.71
    POSITIVE LOGITS
     Dane
    0.78
    ³
    0.75
    itud
    0.73
     Pumpkin
    0.70
    anas
    0.70
    iosity
    0.69
     grandmother
    0.68
    pa
    0.67
     Companion
    0.66
     Prix
    0.65
    Activations Density 2.218%

    No Known Activations