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 768
    5. 8-RES_FS768-JB
    6. 303
    Prev
    Next
    INDEX
    Explanations

    phrases related to protective measures or precautions

    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
    •"make sure"
    "protect"/"security"/"privacy"
    •Make sure feature
    •Canonical Figure 13

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    551
    +0.09
    0.4%
    730
    +0.09
    0.4%
    511
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.09
    0.19
    575
    +0.09
    0.13
    225
    +0.09
    0.11
    Negative Logits
    issance
    -0.80
    zos
    -0.72
    cart
    -0.69
    oric
    -0.69
    union
    -0.68
    ional
    -0.68
    onde
    -0.66
    fred
    -0.64
    Cart
    -0.63
    hower
    -0.60
    POSITIVE LOGITS
    ailability
    0.97
    Protect
    0.87
    ously
    0.87
     perimeter
    0.82
     precautions
    0.82
     Deposit
    0.79
     precaution
    0.79
    egu
    0.78
     against
    0.78
     deposit
    0.76
    Activations Density 5.213%

    No Known Activations