© 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. 1846
    Prev
    Next
    INDEX
    Explanations

    references to historical events and figures

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "the" in various contexts

    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
    tackle
    -0.93
    bnb
    -0.81
    ESPN
    -0.77
    Update
    -0.76
    Reddit
    -0.76
    rame
    -0.76
    ibaba
    -0.76
    leground
    -0.75
    heit
    -0.74
    Downloadha
    -0.74
    POSITIVE LOGITS
     Soviets
    1.31
     earliest
    1.21
     Soviet
    1.19
     USSR
    1.19
     postwar
    1.13
     Nazis
    1.06
     beginnings
    1.03
     latter
    1.03
     Yugoslav
    1.00
     1960
    0.99
    Activations Density 0.677%

    No Known Activations