© 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 768
    9. 8-RES_FS768-JB
    11. 756
    Prev
    Next
    INDEX
    Explanations

    phrases related to political commentary and society opinion

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    501
    +0.10
    0.5%
    203
    +0.10
    0.5%
    449
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    501
    +0.10
    0.13
    203
    +0.10
    0.12
    449
    +0.10
    0.11
    Negative Logits
    ses
    -0.80
    1001
    -0.69
    ãģĨ
    -0.68
    olves
    -0.66
    ema
    -0.63
    iking
    -0.61
    iji
    -0.61
    icc
    -0.61
    sk
    -0.61
    ded
    -0.59
    POSITIVE LOGITS
    romeda
    1.34
     furthermore
    1.18
     secondly
    1.10
     therein
    1.10
    rea
    1.10
     moreover
    1.04
     besides
    1.03
    ersen
    0.98
     lest
    0.94
     yes
    0.94
    Activations Density 2.616%

    No Known Activations