© 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 1536
    9. 8-RES_FS1536-JB
    11. 9
    Prev
    Next
    INDEX
    Explanations

    verbs in the gerund form (ending in 'ing') related to actions and behaviors

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    113
    +0.12
    0.5%
    555
    +0.10
    0.4%
    494
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    113
    +0.12
    0.12
    70
    +0.10
    0.13
    566
    +0.10
    0.11
    Negative Logits
    cia
    -0.70
    croft
    -0.67
    bal
    -0.66
    eka
    -0.65
    phies
    -0.64
    ritical
    -0.63
    peria
    -0.62
    reet
    -0.58
    aun
    -0.58
    sac
    -0.58
    POSITIVE LOGITS
     dangerously
    0.70
     furiously
    0.68
     lately
    0.67
     uphill
    0.67
     pains
    0.66
     frantically
    0.66
    ependent
    0.65
     downhill
    0.64
     trouble
    0.64
     stages
    0.64
    Activations Density 1.990%

    No Known Activations