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 1536
    5. 8-RES_FS1536-JB
    6. 909
    Prev
    Next
    INDEX
    Explanations

    phrases related to policies, regulations, and transparency

    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
    435
    +0.12
    0.5%
    567
    +0.12
    0.5%
    641
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    567
    +0.12
    0.10
    665
    +0.12
    0.16
    435
    +0.11
    0.09
    Negative Logits
     infamous
    -0.70
     strangely
    -0.67
    ties
    -0.66
     inexpl
    -0.65
     oddly
    -0.65
     wonders
    -0.65
     curiously
    -0.63
     speculated
    -0.62
     unlucky
    -0.62
     bizarre
    -0.62
    POSITIVE LOGITS
     ASAP
    1.21
     responsibly
    1.03
     adequate
    0.93
     respectful
    0.91
     unbiased
    0.90
    cknow
    0.90
     properly
    0.90
     truthful
    0.89
     equitable
    0.88
     proper
    0.88
    Activations Density 4.908%

    No Known Activations