© 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 12288
    9. 8-RES_FS12288-JB
    11. 10174
    Prev
    Next
    INDEX
    Explanations

    details related to individuals and their actions or roles

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    430
    +0.13
    0.6%
    154
    +0.10
    0.4%
    115
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    741
    +0.13
    0.05
    115
    +0.10
    0.05
    430
    +0.09
    0.06
    Negative Logits
    olves
    -0.78
    adh
    -0.74
    ¥µ
    -0.65
     outputs
    -0.64
    ellation
    -0.63
     reproduction
    -0.63
    appropriate
    -0.59
    alyses
    -0.59
    akespeare
    -0.58
    itures
    -0.58
    POSITIVE LOGITS
     flanked
    1.08
    enegger
    1.06
     who
    1.06
    whose
    1.05
     whose
    1.03
    who
    1.01
     nicknamed
    0.94
     meanwhile
    0.93
     himself
    0.92
     whom
    0.92
    Activations Density 0.267%

    No Known Activations