© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3332
    Prev
    Next
    INDEX
    Explanations

    substrings following a pattern of slashes and names with varying activation values

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the "REUTERS/" marker indicating news sources

    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
     prol
    -0.77
    worms
    -0.73
     beetles
    -0.72
     worms
    -0.72
     idol
    -0.70
     verbs
    -0.69
     infertility
    -0.68
     sooner
    -0.67
    ĪĴ
    -0.65
     snakes
    -0.65
    POSITIVE LOGITS
    File
    0.99
    AFP
    0.94
    Getty
    0.93
    David
    0.93
    Mike
    0.90
    Jim
    0.89
    Flickr
    0.88
    Rick
    0.87
    Brend
    0.87
    Jonathan
    0.87
    Activations Density 0.026%

    No Known Activations