© 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. 2-RES-JB
    11. 17787
    Prev
    Next
    INDEX
    Explanations

    the surname "Watson."

    oai_token-act-pair · gpt-3.5-turbo

    repeated references to the name "Watson."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ccording
    -0.84
     fertile
    -0.78
    ript
    -0.76
    cised
    -0.75
    eways
    -0.75
    ĻĤ
    -0.74
    ional
    -0.74
    lav
    -0.73
    afort
    -0.72
    ulatory
    -0.71
    POSITIVE LOGITS
     Watson
    1.38
    atson
    1.06
     Holmes
    0.95
     Whe
    0.88
     Hawkins
    0.87
    combe
    0.86
     Cumm
    0.81
     Norton
    0.79
     Evans
    0.78
    ITE
    0.78
    Activations Density 0.006%

    No Known Activations