© 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. 8-RES-JB
    11. 15197
    Prev
    Next
    INDEX
    Explanations

    text patterns starting with special characters followed by words or phrases related to legal or security concepts

    oai_token-act-pair · gpt-3.5-turbo

    repeated use of a specific character or symbol in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     federation
    -0.77
     Negro
    -0.66
     Orchestra
    -0.64
    seless
    -0.63
    iolet
    -0.62
    enium
    -0.62
     regist
    -0.62
     ende
    -0.61
     sacrific
    -0.61
     Federation
    -0.60
    POSITIVE LOGITS
    ï¸ı
    1.02
    ¯
    0.95
    âĢł
    0.92
    cue
    0.85
    STEM
    0.81
    which
    0.76
     namely
    0.75
    cause
    0.74
    #$
    0.73
    °
    0.73
    Activations Density 0.141%

    No Known Activations