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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 768
    5. 8-RES_FS768-JB
    6. 341
    Prev
    Next
    INDEX
    Explanations

    words related to technical processes and methodologies

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    +0.37
    4.7%
    373
    +0.10
    1.3%
    756
    +0.09
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.37
    -0.14
    756
    +0.10
    -0.16
    373
    +0.09
    -0.31
    Negative Logits
    ¶
    -0.21
    )?
    -0.17
     Streaming
    -0.17
     fixme
    -0.16
     [/
    -0.16
    [/
    -0.16
     tonight
    -0.15
     jo
    -0.15
     Posted
    -0.15
     )]
    -0.15
    POSITIVE LOGITS
     advoc
    0.19
    ãĥĥãĥī
    0.19
    acebook
    0.18
    undreds
    0.18
    xtap
    0.18
     acronym
    0.18
    appropriately
    0.18
    agna
    0.18
     enthusi
    0.18
    Enhanced
    0.17
    Activations Density 22.745%

    No Known Activations