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 768
    5. 8-RES_FS768-JB
    6. 593
    Prev
    Next
    INDEX
    Explanations

    personal narratives and anecdotal stories

    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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    450
    +0.12
    0.5%
    287
    +0.10
    0.4%
    241
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    598
    +0.12
    0.10
    287
    +0.10
    0.14
    450
    +0.09
    0.12
    Negative Logits
     guiActiveUnfocused
    -0.69
     ÂŃ
    -0.67
    è¦ļéĨĴ
    -0.63
    ËĪ
    -0.62
    ):
    -0.61
    ],"
    -0.60
    "—
    -0.60
    ")
    -0.60
    ]"
    -0.59
     BRE
    -0.59
    POSITIVE LOGITS
     furthermore
    1.05
     moreover
    0.94
    oret
    0.94
     therefore
    0.93
     downside
    0.88
     however
    0.85
    cknowled
    0.84
    quartered
    0.84
     importantly
    0.84
    anwhile
    0.84
    Activations Density 5.652%

    No Known Activations