© 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. 1019
    Prev
    Next
    INDEX
    Explanations

    phrases related to news content and analysis

    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
    279
    +0.10
    0.5%
    531
    +0.10
    0.4%
    189
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    279
    +0.10
    0.04
    329
    +0.10
    0.03
    13
    +0.10
    0.03
    Negative Logits
    umblr
    -0.76
     Rebell
    -0.71
    aten
    -0.71
    wind
    -0.67
    amazon
    -0.64
    edIn
    -0.63
    atcher
    -0.61
    EStream
    -0.60
     Modes
    -0.60
    abouts
    -0.59
    POSITIVE LOGITS
     insight
    1.13
     assurance
    0.94
     insights
    0.91
     fodder
    0.91
     impetus
    0.88
     valuable
    0.84
     glimpse
    0.83
     rationale
    0.82
     guidance
    0.82
     proof
    0.82
    Activations Density 0.197%

    No Known Activations