© 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. 4-RES-JB
    11. 2902
    Prev
    Next
    INDEX
    Explanations

    acronyms followed by numbers, often in relation to plans or deployments

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific military technology or system related to South Korea

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

    No Comments

    Negative Logits
    ãĤ¹ãĥĪ
    -0.79
    ãĤŃ
    -0.74
    cloth
    -0.71
    keeping
    -0.69
    ãĥ´ãĤ¡
    -0.69
    ãĤ±
    -0.68
    lesi
    -0.68
    atories
    -0.67
    ãĥĥ
    -0.67
    utive
    -0.67
    POSITIVE LOGITS
    OUGH
    1.07
    reads
    0.96
    IRD
    0.93
    ulhu
    0.93
    irteen
    0.91
    ursday
    0.90
    ought
    0.89
    OTAL
    0.88
    READ
    0.87
    urst
    0.87
    Activations Density 0.008%

    No Known Activations