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

    Neuronpedia

    APICircuit 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. 8-RES-JB
    6. 15488
    Prev
    Next
    INDEX
    Explanations

    names or handles of individuals on social media platforms

    oai_token-act-pair · gpt-3.5-turbo

    names of individuals or entities, specifically through social media handles

    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
     captcha
    -0.68
     crystall
    -0.66
     seal
    -0.66
     disarm
    -0.66
     seals
    -0.66
     penal
    -0.65
     warrant
    -0.64
     chilling
    -0.63
     overturn
    -0.63
     humid
    -0.63
    POSITIVE LOGITS
    Jr
    1.08
    _
    1.01
    )
    0.99
    /)
    0.97
    __
    0.93
    Blog
    0.92
    Reports
    0.89
    TX
    0.85
    Studio
    0.85
    @
    0.83
    Activations Density 0.060%

    No Known Activations