© 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. 7356
    Prev
    Next
    INDEX
    Explanations

    phrases starting with the word "Although."

    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
    447
    +0.12
    0.5%
    679
    +0.11
    0.5%
    236
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    236
    +0.12
    0.03
    337
    +0.11
    0.03
    513
    +0.09
    0.03
    Negative Logits
    tnc
    -0.73
    acia
    -0.69
    itto
    -0.65
    bas
    -0.65
    illet
    -0.64
    Jew
    -0.64
    enter
    -0.63
     srfAttach
    -0.61
    taboola
    -0.61
     {*
    -0.60
    POSITIVE LOGITS
     technically
    1.08
    ts
    1.02
     acknowledging
    1.00
     admittedly
    0.90
     conced
    0.85
     initially
    0.84
     ostensibly
    0.78
    tons
    0.76
     researching
    0.75
     outnumbered
    0.72
    Activations Density 0.047%

    No Known Activations