© 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. 7274
    Prev
    Next
    INDEX
    Explanations

    mentions of licenses or licensing in the text

    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
    459
    +0.13
    0.6%
    508
    +0.12
    0.6%
    168
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    508
    +0.13
    0.02
    575
    +0.12
    0.02
    168
    +0.11
    0.02
    Negative Logits
     Lans
    -0.96
    ilon
    -0.77
    te
    -0.69
     IMAGES
    -0.68
    telling
    -0.67
    dain
    -0.66
    ahi
    -0.66
     Remem
    -0.65
    terson
    -0.64
    ilight
    -0.63
    POSITIVE LOGITS
     licenses
    1.07
     licensing
    0.95
    Reviewer
    0.94
    license
    0.93
     licences
    0.90
     waivers
    0.89
     license
    0.88
     agreements
    0.87
    enses
    0.86
     renewal
    0.86
    Activations Density 0.025%

    No Known Activations