© 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. 671
    Prev
    Next
    INDEX
    Explanations

    the preposition 'to' followed by a verb

    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
    608
    +0.18
    0.8%
    718
    +0.17
    0.8%
    104
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.18
    0.24
    718
    +0.17
    0.21
    49
    +0.10
    0.17
    Negative Logits
    Got
    -0.64
     wastes
    -0.62
    calling
    -0.61
    expected
    -0.56
     ain
    -0.56
     copied
    -0.55
     packets
    -0.55
    need
    -0.54
     rolled
    -0.54
     flagged
    -0.54
    POSITIVE LOGITS
    ggles
    1.40
     ensure
    1.34
     satisfy
    1.31
     compensate
    1.28
     accommodate
    1.26
     avoid
    1.26
     maximize
    1.26
    ilet
    1.25
     commemorate
    1.22
     facilitate
    1.22
    Activations Density 2.318%

    No Known Activations