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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 407
    Prev
    Next
    INDEX
    Explanations

    terms related to legal processes and negotiations

    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
    749
    +0.15
    0.7%
    333
    +0.11
    0.5%
    113
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    749
    +0.15
    0.12
    55
    +0.11
    0.13
    531
    +0.10
    0.09
    Negative Logits
     candy
    -0.68
     representations
    -0.64
    oak
    -0.62
     clutter
    -0.62
     Sul
    -0.60
     abusers
    -0.59
    stadt
    -0.59
    lder
    -0.58
    Reading
    -0.58
     liar
    -0.56
    POSITIVE LOGITS
     underway
    1.50
     undertaken
    1.28
     conducted
    1.23
     occurring
    1.11
     commenced
    1.07
     occur
    1.04
     initiated
    1.03
     involving
    1.03
     begun
    1.01
     happening
    1.00
    Activations Density 2.730%

    No Known Activations