© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 1732
    Prev
    Next
    INDEX
    Explanations

    the word "busy" or phrases indicating busyness and activity

    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
    726
    +0.12
    0.6%
    471
    +0.11
    0.5%
    760
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    726
    +0.12
    0.02
    111
    +0.11
    0.01
    271
    +0.10
    0.01
    Negative Logits
    otide
    -0.69
    INAL
    -0.69
    Merit
    -0.68
    uer
    -0.63
    âĵĺ
    -0.62
     resemblance
    -0.61
    IZE
    -0.60
    ulner
    -0.60
    uana
    -0.60
    arium
    -0.60
    POSITIVE LOGITS
    ming
    0.93
     schedules
    0.84
    cler
    0.83
    bus
    0.83
    body
    0.81
     juggling
    0.79
    work
    0.79
    ness
    0.76
    working
    0.76
    Corp
    0.74
    Activations Density 0.019%

    No Known Activations