© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 10335
    Prev
    Next
    INDEX
    Explanations

    expressions related to recognition, acknowledgment, and appreciation

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    36
    +0.13
    0.6%
    374
    +0.11
    0.5%
    94
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.13
    0.09
    636
    +0.11
    0.06
    36
    +0.11
    0.07
    Negative Logits
    SPONSORED
    -1.11
    enced
    -0.95
    yll
    -0.91
    fulness
    -0.77
    pps
    -0.76
    ocumented
    -0.75
    erson
    -0.74
    pires
    -0.71
    pired
    -0.70
    assador
    -0.70
    POSITIVE LOGITS
     basics
    0.98
     slack
    0.88
     mantle
    0.85
     cylinders
    0.84
     seams
    0.83
     chores
    0.82
     reins
    0.81
     towel
    0.81
     weeds
    0.79
     goodies
    0.78
    Activations Density 0.664%

    No Known Activations