© 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. 1744
    Prev
    Next
    INDEX
    Explanations

    phrases related to joining or signing up for actions or programs

    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
    767
    +0.09
    0.4%
    154
    +0.09
    0.4%
    750
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    194
    +0.09
    0.02
    74
    +0.09
    0.02
    680
    +0.09
    0.02
    Negative Logits
    estamp
    -0.65
    VALUE
    -0.63
    ometry
    -0.62
    omorphic
    -0.62
    ufact
    -0.61
    aeda
    -0.60
     disple
    -0.59
    Origin
    -0.58
    omorph
    -0.57
    bara
    -0.57
    POSITIVE LOGITS
    blogs
    0.75
     Forbes
    0.73
     podcasts
    0.72
     archive
    0.70
    PHOTOS
    0.69
     Subscribe
    0.69
    Politics
    0.69
    aily
    0.67
    articles
    0.67
    Sources
    0.67
    Activations Density 0.076%

    No Known Activations