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 3072
    5. 8-RES_FS3072-JB
    6. 419
    Prev
    Next
    INDEX
    Explanations

    information about upcoming releases or events

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    87
    +0.24
    1.2%
    435
    +0.10
    0.5%
    665
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.24
    0.18
    526
    +0.10
    -0.04
    87
    +0.09
    -0.09
    Negative Logits
    @#&
    -0.57
    verbs
    -0.56
    fortunately
    -0.54
     wonderfully
    -0.53
     beautifully
    -0.51
     hilar
    -0.50
     fortunately
    -0.49
     strangely
    -0.48
     Sina
    -0.48
     reminds
    -0.47
    POSITIVE LOGITS
     someday
    1.32
     anytime
    0.98
     sooner
    0.98
     future
    0.93
     ASAP
    0.88
     next
    0.86
     tomorrow
    0.86
     2020
    0.82
     forthcoming
    0.79
     hereafter
    0.78
    Activations Density 5.460%

    No Known Activations