© 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. 11-RES-JB
    11. 18669
    Prev
    Next
    INDEX
    Explanations

    phrases related to consent and options, typically found in terms and conditions or subscription forms

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to subscription options and user control

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inav
    -0.62
    omorph
    -0.61
     suspic
    -0.55
    elled
    -0.55
    alam
    -0.54
     alot
    -0.53
    ãĥ¥
    -0.53
    isers
    -0.53
     quart
    -0.52
    cade
    -0.51
    POSITIVE LOGITS
    void
    0.65
    ascript
    0.63
    enter
    0.60
    Newsletter
    0.60
     ILCS
    0.59
    odder
    0.58
    idav
    0.57
    .
    0.56
    llular
    0.56
    BLIC
    0.55
    Activations Density 0.018%

    No Known Activations