Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 18117
    Prev
    Next
    INDEX
    Explanations

    mentions of terrorist groups such as Al Qaeda, in particular, the name "Al" followed by a high activation value

    oai_token-act-pair · gpt-3.5-turbo

    references to "Al Qaeda."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãĥ¼ãĥĨãĤ£
    -0.93
     Vaugh
    -0.83
     Tycoon
    -0.78
    manship
    -0.76
     Cricket
    -0.75
    å£
    -0.73
     Mechdragon
    -0.72
     Lovecraft
    -0.71
    lihood
    -0.70
     Seah
    -0.70
    POSITIVE LOGITS
    manac
    1.14
     Jazeera
    1.13
    gebra
    1.11
    addin
    1.07
    ameda
    1.00
    umni
    0.98
     Qaeda
    0.97
    iyah
    0.96
    oha
    0.96
    una
    0.94
    Activations Density 0.020%

    No Known Activations