© 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. 1-RES-JB
    11. 3318
    Prev
    Next
    INDEX
    Explanations

    complex composed characters or character combinations

    oai_token-act-pair · gpt-3.5-turbo

    sequences of obscure characters or symbols that may indicate specialized content or encoding

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    xus
    -0.97
    lycer
    -0.95
    oche
    -0.87
    olit
    -0.82
    ffen
    -0.82
    orne
    -0.80
    ulously
    -0.79
    atche
    -0.78
    tera
    -0.77
    eways
    -0.77
    POSITIVE LOGITS
    ãģ¦
    1.75
    ãģĦ
    1.68
    ãĤĭ
    1.60
    ãģ
    1.56
    ãģŁ
    1.53
    ãģ¾
    1.51
    ãĤĵ
    1.49
    ãģª
    1.47
    ãĤī
    1.46
    ãĤ
    1.42
    Activations Density 0.011%

    No Known Activations