© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3-RES-JB
    11. 5833
    Prev
    Next
    INDEX
    Explanations

    instances where investigators are mentioned in a text

    oai_token-act-pair · gpt-3.5-turbo

    references to investigators involved in various cases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yip
    -0.75
    ward
    -0.67
    ãĤª
    -0.65
    oted
    -0.65
    cock
    -0.65
    ned
    -0.64
    nz
    -0.64
    sup
    -0.63
    bows
    -0.61
    rip
    -0.60
    POSITIVE LOGITS
     investigator
    0.99
     investigating
    0.95
     probing
    0.92
     subpoen
    0.86
     uncovered
    0.85
     investigation
    0.83
     investigators
    0.83
     investigates
    0.79
    nsics
    0.78
     investigate
    0.78
    Activations Density 0.018%

    No Known Activations