© 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. 10-RES-JB
    11. 6732
    Prev
    Next
    INDEX
    Explanations

    phrases related to thorough searching or exploration

    oai_token-act-pair · gpt-3.5-turbo

    instances of searching or navigating through something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ajor
    -0.85
    ESA
    -0.79
    nown
    -0.79
    roit
    -0.75
    ifer
    -0.74
    immer
    -0.73
    orthy
    -0.72
    lihood
    -0.72
    ablishment
    -0.71
    Enlarge
    -0.70
    POSITIVE LOGITS
     crappy
    1.27
     shitty
    1.24
     crap
    1.24
     stuff
    1.15
     shit
    1.13
     stupid
    1.06
     dudes
    1.03
     boobs
    1.02
     gadgets
    1.02
     idiots
    1.01
    Activations Density 0.448%

    No Known Activations