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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 7-RES-JB
    6. 17741
    Prev
    Next
    INDEX
    Explanations

    phrases beginning with "The", often at the start of a sentence or headline

    oai_token-act-pair · gpt-3.5-turbo

    end-of-text markers or signals indicating the conclusion of a document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ÏĢ
    -0.83
    Ïī
    -0.76
    .","
    -0.71
    !.
    -0.71
    ceive
    -0.69
    ����
    -0.69
    cape
    -0.69
    .</
    -0.69
     ("
    -0.68
    gpu
    -0.67
    POSITIVE LOGITS
    resa
    1.56
    odore
    1.49
    oret
    1.33
    ories
    1.17
     latest
    1.08
     irony
    1.04
     latter
    1.01
     biggest
    0.96
     remainder
    0.94
    nce
    0.92
    Activations Density 0.320%

    No Known Activations