© 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. 1908
    Prev
    Next
    INDEX
    Explanations

    references to specific individuals, often describing their actions or roles

    oai_token-act-pair · gpt-3.5-turbo

    references to specific individuals and their actions or characteristics

    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
    Services
    -0.66
    Repeat
    -0.61
    Herm
    -0.60
     weakening
    -0.60
     IRA
    -0.60
    Adjust
    -0.59
     PP
    -0.58
    services
    -0.58
     compromising
    -0.57
     sleeper
    -0.57
    POSITIVE LOGITS
    pherd
    1.12
    zbollah
    0.95
    'll
    0.94
    reditary
    0.94
     joins
    0.91
    ffield
    0.90
    'd
    0.90
     wrote
    0.86
    enegger
    0.84
     oversees
    0.83
    Activations Density 0.307%

    No Known Activations