© 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. 1800
    Prev
    Next
    INDEX
    Explanations

    keywords related to additional information or related content, such as "Also read" or "Also on."

    oai_token-act-pair · gpt-3.5-turbo

    repeated phrases that indicate additional information or references

    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
    eers
    -0.68
    anse
    -0.67
    hoe
    -0.67
    slot
    -0.67
    clock
    -0.66
    mission
    -0.65
    jam
    -0.64
     Rampage
    -0.62
    aciously
    -0.59
    acious
    -0.59
    POSITIVE LOGITS
     noteworthy
    0.93
     Known
    0.86
     notable
    0.82
     included
    0.76
     notor
    0.75
     readable
    0.74
     includ
    0.72
     applicable
    0.72
     referred
    0.72
     Called
    0.71
    Activations Density 0.067%

    No Known Activations