© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 7746
    Prev
    Next
    INDEX
    Explanations

    adjectives and phrases indicating a strong or emphatic emphasis

    oai_token-act-pair · gpt-3.5-turbo

    terms indicating strong success or dominant support

    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
    emis
    -0.86
    paces
    -0.85
    pace
    -0.80
    ecast
    -0.72
    azine
    -0.72
    natureconservancy
    -0.71
    cffff
    -0.70
    reenshots
    -0.69
    rosso
    -0.69
    uses
    -0.67
    POSITIVE LOGITS
     YES
    0.94
     rejection
    0.88
     majority
    0.87
     majorities
    0.86
     negativity
    0.81
     dominance
    0.79
     condemnation
    0.78
     refusal
    0.76
     necessity
    0.75
     vict
    0.75
    Activations Density 0.120%

    No Known Activations