© 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. 6285
    Prev
    Next
    INDEX
    Explanations

    specific terms related to political events and figures, particularly in investigations or matters of authority

    oai_token-act-pair · gpt-3.5-turbo

    references to political figures and influence

    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
    cknowled
    -0.64
    oneliness
    -0.53
     nutshell
    -0.52
    pher
    -0.52
    arij
    -0.51
     consisted
    -0.50
    itsch
    -0.49
     Substance
    -0.49
     Bicycle
    -0.48
    mma
    -0.47
    POSITIVE LOGITS
     anew
    0.59
    )).
    0.57
    DERR
    0.56
    onto
    0.55
    ifully
    0.54
     differently
    0.53
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.53
     quotas
    0.53
     anymore
    0.53
    matically
    0.53
    Activations Density 2.804%

    No Known Activations