© 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. 1645
    Prev
    Next
    INDEX
    Explanations

    phrases related to being at the forefront of an industry or technology

    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
    181
    +0.14
    0.7%
    447
    +0.14
    0.6%
    554
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    737
    +0.14
    0.03
    207
    +0.14
    0.02
    197
    +0.11
    0.03
    Negative Logits
     disemb
    -0.80
     stub
    -0.71
    rafted
    -0.66
     Wolves
    -0.63
    ©¶æ
    -0.62
    juries
    -0.61
     Tsukuyomi
    -0.60
    ires
    -0.59
     Pengu
    -0.59
    iannopoulos
    -0.59
    POSITIVE LOGITS
     forefront
    0.86
     innov
    0.85
    selling
    0.83
     technological
    0.79
     worldwide
    0.78
     globally
    0.76
     innovation
    0.75
    boards
    0.73
     amongst
    0.72
     supplying
    0.72
    Activations Density 0.119%

    No Known Activations