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. 23098
    Prev
    Next
    INDEX
    Explanations

    terms related to suspension and expulsion in educational and legal contexts

    oai_token-act-pair · gpt-4o-miniTriggered 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
    274
    +0.13
    0.6%
    718
    +0.11
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    301
    +0.13
    0.02
    274
    +0.11
    0.02
    189
    +0.10
    0.01
    Negative Logits
    rics
    -0.79
    ipped
    -0.76
    zig
    -0.76
    ebus
    -0.74
    illard
    -0.71
    åĮ
    -0.70
    opher
    -0.69
    icians
    -0.67
     originals
    -0.67
    swick
    -0.66
    POSITIVE LOGITS
     disbelief
    1.18
     indefinitely
    1.15
     suspensions
    0.91
     suspending
    0.91
     suspended
    0.89
     suspension
    0.87
    ishment
    0.84
     sentences
    0.75
     suspend
    0.75
    icious
    0.73
    Activations Density 0.024%

    No Known Activations