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. 1175
    Prev
    Next
    INDEX
    Explanations

    information related to research data and scientific findings in various 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
    417
    +0.11
    0.5%
    749
    +0.10
    0.5%
    715
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.11
    0.05
    647
    +0.10
    0.06
    749
    +0.10
    0.06
    Negative Logits
    thing
    -0.84
    stones
    -0.82
    aries
    -0.75
    mates
    -0.74
    gets
    -0.70
    atom
    -0.69
    ices
    -0.68
    matter
    -0.68
    woods
    -0.66
    shake
    -0.66
    POSITIVE LOGITS
     efficiency
    1.17
     visibility
    1.15
     likelihood
    1.13
     susceptibility
    1.08
     sensitivity
    1.06
     responsiveness
    1.04
     competitiveness
    1.02
     awareness
    1.02
     frequency
    1.01
     productivity
    1.01
    Activations Density 0.750%

    No Known Activations