© 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 3072
    9. 8-RES_FS3072-JB
    11. 3034
    Prev
    Next
    INDEX
    Explanations

    comments or dialogue related to controversial or sensitive topics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    762
    +0.12
    0.6%
    737
    +0.10
    0.5%
    481
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.12
    0.18
    393
    +0.10
    0.18
    351
    +0.09
    0.08
    Negative Logits
    ãĢij
    -0.72
     ®
    -0.66
     Alas
    -0.65
     Released
    -0.65
    ãĤ¹ãĥĪ
    -0.64
     âī¡
    -0.63
     eagerly
    -0.62
     Suff
    -0.61
    %%
    -0.61
     Amid
    -0.60
    POSITIVE LOGITS
     everybody
    1.10
     gonna
    1.10
     [
    1.10
     uh
    1.08
     somebody
    1.08
     yeah
    1.06
     kinda
    1.02
     basically
    1.01
    entimes
    0.99
     kind
    0.99
    Activations Density 3.759%

    No Known Activations