Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 968
    Prev
    Next
    INDEX
    Explanations

    phrases related to board games

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    288
    +0.12
    0.5%
    737
    +0.11
    0.5%
    582
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.17
    325
    +0.11
    0.12
    678
    +0.10
    0.09
    Negative Logits
     Uz
    -0.75
    BAT
    -0.72
    ETH
    -0.71
     righteousness
    -0.68
    alez
    -0.67
    DIT
    -0.66
     Hots
    -0.66
     CBI
    -0.64
     Jarrett
    -0.63
     Vegas
    -0.62
    POSITIVE LOGITS
    tops
    1.30
    chair
    1.03
    room
    0.99
    chairs
    0.97
    walk
    0.97
    mates
    0.96
    cloth
    0.94
    top
    0.90
    holder
    0.87
    builders
    0.85
    Activations Density 2.529%

    No Known Activations