© 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. 6863
    Prev
    Next
    INDEX
    Explanations

    challenges or tasks stated clearly and precisely

    oai_token-act-pair · gpt-3.5-turbo

    expressions of inquiries or challenges presented in a conversational context

    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
     Romantic
    -0.68
     ageing
    -0.65
     trendy
    -0.65
    æ©
    -0.65
     charms
    -0.64
    atars
    -0.64
    ãĤ¼
    -0.64
     lure
    -0.64
     lured
    -0.63
     Equip
    -0.63
    POSITIVE LOGITS
     briefed
    1.09
     redacted
    1.01
     confidentiality
    0.94
     subpoen
    0.94
     briefings
    0.93
     subpoena
    0.93
     prosecut
    0.90
     jurisd
    0.90
     sir
    0.90
     confidential
    0.89
    Activations Density 0.972%

    No Known Activations