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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 1575
    Prev
    Next
    INDEX
    Explanations

    titles or mentions of notable individuals in the political or technological fields

    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
    393
    +0.12
    0.5%
    463
    +0.11
    0.5%
    391
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    463
    +0.12
    0.07
    393
    +0.11
    0.08
    703
    +0.09
    0.05
    Negative Logits
    ETS
    -0.84
    pees
    -0.81
    ciating
    -0.80
    idays
    -0.79
    days
    -0.79
    orrow
    -0.78
    ences
    -0.78
    zees
    -0.77
    oks
    -0.77
    ents
    -0.77
    POSITIVE LOGITS
     educator
    1.19
     translator
    1.13
     lecturer
    1.07
     commentator
    1.07
     entrepreneur
    1.07
     longtime
    1.07
     thinker
    1.05
     namesake
    1.05
     supporter
    1.04
     former
    1.03
    Activations Density 0.887%

    No Known Activations