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. 15300
    Prev
    Next
    INDEX
    Explanations

    adjective-noun combinations where the adjective describes the importance or necessity of the noun

    oai_token-act-pair · gpt-3.5-turbo

    terms related to significance and essentiality

    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
    events
    -0.90
    ancies
    -0.87
    iencies
    -0.86
    bows
    -0.83
    apses
    -0.83
    onies
    -0.81
    lees
    -0.81
    ptions
    -0.79
    atism
    -0.78
    rooms
    -0.77
    POSITIVE LOGITS
     contributor
    1.29
     indicator
    1.23
     reminder
    1.19
     contender
    1.19
     source
    1.17
     deterrent
    1.13
     example
    1.13
     fixture
    1.09
     indication
    1.09
     predictor
    1.08
    Activations Density 0.188%

    No Known Activations