© 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. 1223
    Prev
    Next
    INDEX
    Explanations

    mentions of commercial activities or products

    oai_token-act-pair · gpt-3.5-turbo

    references to commercial activities and products

    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
     Utt
    -0.77
     Held
    -0.73
    hid
    -0.69
     Khe
    -0.68
    toe
    -0.67
    lvl
    -0.65
    enum
    -0.64
    guards
    -0.64
     Beasts
    -0.64
     comrade
    -0.63
    POSITIVE LOGITS
     commercial
    3.65
    Commercial
    2.91
    commercial
    2.64
     Commercial
    2.63
     commer
    2.47
     commercially
    2.41
    mercial
    1.93
     commercials
    1.64
     industrial
    1.42
     commerce
    1.30
    Activations Density 0.018%

    No Known Activations