© 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 24576
    9. 8-RES_FS24576-JB
    11. 20030
    Prev
    Next
    INDEX
    Explanations

    time-related phrases and dates

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    218
    +0.10
    0.5%
    283
    +0.09
    0.4%
    84
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    98
    +0.10
    0.06
    702
    +0.09
    0.04
    677
    +0.09
    0.04
    Negative Logits
    Reviewer
    -0.71
     wrongly
    -0.68
     summed
    -0.66
     aloud
    -0.61
     raged
    -0.61
     outnumbered
    -0.60
     alike
    -0.59
     sear
    -0.59
     unwittingly
    -0.59
     hurled
    -0.59
    POSITIVE LOGITS
     2019
    1.40
     2020
    1.25
     2021
    1.25
    2019
    1.21
     2018
    1.08
    2020
    1.03
     TBA
    1.03
    uesday
    1.02
     2024
    0.99
     2022
    0.98
    Activations Density 0.318%

    No Known Activations