© 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. 9284
    Prev
    Next
    INDEX
    Explanations

    sentences with a strong emotional or opinionated tone

    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
    221
    +0.11
    0.5%
    630
    +0.11
    0.5%
    218
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    455
    +0.11
    0.04
    620
    +0.11
    0.02
    718
    +0.10
    0.04
    Negative Logits
     sometime
    -0.72
    itton
    -0.65
     Opportun
    -0.64
    uctions
    -0.63
    ortment
    -0.63
    ahime
    -0.62
     Locations
    -0.61
    iland
    -0.58
     Targ
    -0.58
     Pars
    -0.57
    POSITIVE LOGITS
     nor
    1.39
     whatsoever
    1.31
     ever
    1.27
     EVER
    1.25
     necessarily
    1.21
     anymore
    1.16
    nor
    1.06
     bothered
    0.94
    except
    0.94
     slightest
    0.88
    Activations Density 0.168%

    No Known Activations