© 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 1536
    9. 8-RES_FS1536-JB
    11. 31
    Prev
    Next
    INDEX
    Explanations

    phrases related to conservative political perspectives and figures

    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
    288
    +0.15
    0.7%
    18
    +0.14
    0.6%
    447
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.18
    18
    +0.14
    0.13
    88
    +0.11
    0.08
    Negative Logits
     Drawn
    -0.77
    ĸļ
    -0.76
    upon
    -0.75
    phies
    -0.72
    ptin
    -0.68
    ividually
    -0.67
    drops
    -0.67
    ciating
    -0.67
    usable
    -0.67
    ilage
    -0.67
    POSITIVE LOGITS
     tendencies
    1.24
    ism
    1.14
     sensibilities
    1.10
     ideals
    1.07
     chic
    1.05
     minded
    1.03
     ideology
    1.01
     zeal
    0.99
     thinkers
    0.95
     mindset
    0.94
    Activations Density 2.280%

    No Known Activations