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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 650
    Prev
    Next
    INDEX
    Explanations

    dates of the week

    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
    430
    +0.20
    0.9%
    384
    +0.11
    0.5%
    662
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.20
    0.15
    384
    +0.11
    0.10
    662
    +0.10
    0.11
    Negative Logits
    åĬ
    -0.73
    respons
    -0.71
    egal
    -0.68
    ript
    -0.66
    ership
    -0.66
    bler
    -0.65
    abet
    -0.65
    chest
    -0.64
    ordes
    -0.64
    onne
    -0.64
    POSITIVE LOGITS
     afternoon
    1.80
     morning
    1.75
     evening
    1.66
     night
    1.60
     mornings
    1.45
    morning
    1.27
     Night
    1.24
     nights
    1.22
     evenings
    1.20
    night
    1.05
    Activations Density 0.710%

    No Known Activations