Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 209
    Prev
    Next
    INDEX
    Explanations

    language related to skepticism towards scientific consensus, government agencies, universities, and medical research related to global warming and public health

    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
    480
    +0.43
    2.1%
    4
    +0.11
    0.5%
    351
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.43
    -0.00
    351
    +0.11
    0.06
    526
    +0.10
    -0.01
    Negative Logits
     Compass
    -0.67
     Finder
    -0.66
    TPPStreamerBot
    -0.65
    accompan
    -0.65
    REE
    -0.65
     Osw
    -0.63
     Observ
    -0.62
     Companion
    -0.62
     Fres
    -0.61
    availability
    -0.61
    POSITIVE LOGITS
     scapego
    1.05
     extortion
    1.00
     impunity
    0.99
     blackmail
    0.97
     phony
    0.96
     bogus
    0.95
     shameless
    0.91
     monopol
    0.90
     gull
    0.90
     unsuspecting
    0.90
    Activations Density 1.654%

    No Known Activations