© 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. 9301
    Prev
    Next
    INDEX
    Explanations

    phrases related to personal opinions or thoughts

    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
    532
    +0.10
    0.5%
    380
    +0.09
    0.4%
    634
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    380
    +0.10
    0.02
    395
    +0.09
    0.02
    532
    +0.09
    0.02
    Negative Logits
    ucket
    -0.88
    \/\/
    -0.75
    ãĥł
    -0.74
    abase
    -0.74
    thumbnails
    -0.73
    ccording
    -0.73
    iant
    -0.71
    entin
    -0.69
    teness
    -0.68
    hyd
    -0.68
    POSITIVE LOGITS
     voices
    1.21
     loud
    1.18
     roar
    1.15
     firsthand
    1.13
     voice
    1.04
     noises
    1.03
     cries
    1.01
     footsteps
    1.01
     whispers
    0.98
     Voices
    0.97
    Activations Density 0.085%

    No Known Activations