© 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. 16589
    Prev
    Next
    INDEX
    Explanations

    mentions of the name "Parsons"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific individuals, particularly in sports and entertainment contexts

    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
     bend
    -0.70
    maid
    -0.69
    station
    -0.68
    ournals
    -0.67
     Zen
    -0.66
    iband
    -0.66
     upstream
    -0.66
    upuncture
    -0.65
    ivo
    -0.65
    ocard
    -0.65
    POSITIVE LOGITS
     Parsons
    2.36
     Trevor
    1.57
     Cul
    1.56
     Lub
    1.29
     Turing
    1.26
     Larson
    1.14
     Kelley
    1.07
     gunman
    1.04
     Chaff
    1.03
     Pratt
    1.01
    Activations Density 0.063%

    No Known Activations