© 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. 7169
    Prev
    Next
    INDEX
    Explanations

    locations and events related to incidents or disasters

    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
    145
    +0.10
    0.4%
    99
    +0.10
    0.4%
    184
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    145
    +0.10
    0.05
    232
    +0.10
    0.05
    24
    +0.09
    0.04
    Negative Logits
    agonists
    -0.75
    icter
    -0.65
    obin
    -0.65
     carbohyd
    -0.64
    PDATE
    -0.64
    ritical
    -0.63
    umsy
    -0.63
    odore
    -0.59
     rul
    -0.59
    aeper
    -0.59
    POSITIVE LOGITS
     Resort
    1.23
     Parish
    1.14
     Cemetery
    1.00
     LLP
    1.00
     Festival
    0.93
     Hotel
    0.92
     Ltd
    0.90
     Foundation
    0.90
     Estate
    0.90
     Massacre
    0.88
    Activations Density 0.231%

    No Known Activations