© 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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 6814
    Prev
    Next
    INDEX
    Explanations

    the word "somewhat."

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "somewhere."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PLA
    -0.69
     Runner
    -0.68
    Office
    -0.68
     Fighters
    -0.68
     Whitney
    -0.65
    cells
    -0.65
     Slovenia
    -0.65
     Muller
    -0.64
     Cell
    -0.64
     KB
    -0.63
    POSITIVE LOGITS
    terness
    1.03
    eteen
    1.02
    eteenth
    0.96
    elfth
    0.94
    eenth
    0.94
    cipled
    0.92
    ingu
    0.90
    iculty
    0.87
    rontal
    0.85
    inguished
    0.85
    Activations Density 0.008%

    No Known Activations