© 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. 17007
    Prev
    Next
    INDEX
    Explanations

    terms related to open-source software and projects

    oai_token-act-pair · gpt-3.5-turbo

    terms related to open-source technology and web security

    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
    •List
    It's funny as hell

    No Comments

    Negative Logits
    ravings
    -0.84
    lees
    -0.78
    poons
    -0.76
    Ire
    -0.76
    archs
    -0.76
    aters
    -0.75
    SPONSORED
    -0.75
    clips
    -0.74
    rants
    -0.73
    idents
    -0.73
    POSITIVE LOGITS
     acronym
    1.08
     abbre
    0.96
     successor
    0.95
     entity
    0.94
     powerhouse
    0.90
     version
    0.89
     tool
    0.88
     phenomenon
    0.87
     variant
    0.86
     organization
    0.86
    Activations Density 0.403%

    No Known Activations