© 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. 20228
    Prev
    Next
    INDEX
    Explanations

    HTML and programming code elements in text

    oai_token-act-pair · gpt-3.5-turbo

    social media handles or mentions

    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
     mosquito
    -0.52
     influenza
    -0.50
     pollen
    -0.50
     GMO
    -0.49
     filament
    -0.49
     gravitational
    -0.49
     odor
    -0.49
     tire
    -0.49
     measles
    -0.48
     approximate
    -0.48
    POSITIVE LOGITS
    eele
    0.60
    anwhile
    0.59
    ourning
    0.58
    ogether
    0.57
    sequently
    0.56
    nce
    0.56
    flix
    0.55
    prisingly
    0.55
    ueless
    0.54
    rison
    0.53
    Activations Density 1.148%

    No Known Activations