© 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. 899
    Prev
    Next
    INDEX
    Explanations

    information related to nutrition and calories

    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
    746
    +0.11
    0.5%
    146
    +0.11
    0.5%
    28
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.11
    0.05
    288
    +0.11
    0.06
    57
    +0.10
    0.03
    Negative Logits
    ioned
    -1.11
    ions
    -1.01
    ership
    -1.00
    ivity
    -0.87
    ansas
    -0.86
    oan
    -0.85
    enic
    -0.84
    enstein
    -0.84
    nesses
    -0.83
    essee
    -0.82
    POSITIVE LOGITS
    Ø©
    0.90
     Clarke
    0.71
     Flames
    0.65
     Sparks
    0.64
     srfAttach
    0.63
    Frames
    0.62
     Schmidt
    0.62
     Manson
    0.61
     fries
    0.61
     Libre
    0.60
    Activations Density 0.270%

    No Known Activations