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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 43274
    Prev
    Next
    INDEX
    Explanations

    phrases and numbers related to promotional offers or deals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    579
    +0.11
    0.5%
    592
    +0.11
    0.5%
    109
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    592
    +0.11
    0.01
    474
    +0.11
    0.01
    666
    +0.10
    0.01
    Negative Logits
     misinterpret
    -0.70
    cycl
    -0.65
     grop
    -0.65
    amination
    -0.63
    umably
    -0.63
    grass
    -0.61
     unaccount
    -0.61
     pil
    -0.60
     bicycl
    -0.60
     conjecture
    -0.60
    POSITIVE LOGITS
     Subscribe
    0.91
     subscribing
    0.87
    Learn
    0.86
     subscribe
    0.80
     Achieve
    0.80
     Learn
    0.79
    ifestyle
    0.76
     Membership
    0.74
     Visit
    0.73
     subscription
    0.73
    Activations Density 0.016%

    No Known Activations