© 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. 18513
    Prev
    Next
    INDEX
    Explanations

    phrases or sequences of words related to proper nouns such as movie titles, place names, and technology terms

    oai_token-act-pair · gpt-3.5-turbo

    sentences or phrases that signify a concluding statement

    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
    ozyg
    -0.79
     involuntary
    -0.66
    thouse
    -0.65
     advis
    -0.64
     oak
    -0.64
    inous
    -0.62
     idle
    -0.62
     bullied
    -0.61
     unch
    -0.61
     barric
    -0.61
    POSITIVE LOGITS
     respectively
    1.30
     These
    1.29
     Both
    1.23
     Each
    1.23
     Together
    1.21
     Also
    1.19
     They
    1.19
     Basically
    1.17
     Additionally
    1.12
     Those
    1.10
    Activations Density 0.540%

    No Known Activations