© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 1536
    9. 8-RES_FS1536-JB
    11. 1141
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to police investigations and legal actions

    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
    447
    +0.17
    0.8%
    363
    +0.13
    0.6%
    346
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    346
    +0.17
    0.14
    363
    +0.13
    0.11
    281
    +0.11
    0.08
    Negative Logits
    ":"/
    -0.85
    except
    -0.76
    respective
    -0.75
    ":["
    -0.74
     accordingly
    -0.70
    tarians
    -0.70
     depends
    -0.69
    inct
    -0.69
    olars
    -0.67
     essentials
    -0.66
    POSITIVE LOGITS
     accidentally
    1.02
     mistakenly
    0.96
     overheard
    0.91
     recently
    0.80
     allegedly
    0.80
    igslist
    0.78
     inadvertently
    0.77
    iphany
    0.77
     unexpectedly
    0.76
     classmate
    0.72
    Activations Density 6.848%

    No Known Activations