© 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. 4-RES-JB
    11. 10289
    Prev
    Next
    INDEX
    Explanations

    proper names or names of people, potentially focusing on a specific name "Fran"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific names, particularly "Fran" and "Cait"

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

    No Comments

    Negative Logits
    eenth
    -0.80
    wrapper
    -0.70
    gow
    -0.68
    wagen
    -0.68
     Graveyard
    -0.67
    orld
    -0.66
    ylum
    -0.66
    andr
    -0.66
    ivid
    -0.65
    ambers
    -0.65
    POSITIVE LOGITS
    kel
    1.07
     Dres
    1.04
     Fran
    0.97
    zen
    0.93
    furt
    0.89
    thal
    0.85
    ck
    0.83
    cium
    0.81
    tha
    0.79
    rier
    0.78
    Activations Density 0.008%

    No Known Activations