© 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. 0-RES-JB
    11. 22211
    Prev
    Next
    INDEX
    Explanations

    phrases related to a specific individual named "Bundy"

    oai_token-act-pair · gpt-3.5-turbo

    references to the Bundy family and related ranching issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hran
    -0.81
     Io
    -0.78
    hend
    -0.76
     analyse
    -0.76
    ENG
    -0.75
     Bok
    -0.72
     drinks
    -0.72
     Aki
    -0.71
    aft
    -0.70
    anish
    -0.70
    POSITIVE LOGITS
     Bundy
    3.89
     ranc
    2.23
     BLM
    1.67
     ranch
    1.64
     wildfire
    1.53
     Reno
    1.41
     Manson
    1.24
     Walmart
    1.23
     wildfires
    1.21
     grazing
    1.17
    Activations Density 0.036%

    No Known Activations