© 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. 7961
    Prev
    Next
    INDEX
    Explanations

    terms related to the role or importance of something

    oai_token-act-pair · gpt-3.5-turbo

    phrases concerning the concept of "role" in various contexts

    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
     Pwr
    -0.84
    Catalog
    -0.82
    burst
    -0.80
    igue
    -0.73
     è£ıè¦ļéĨĴ
    -0.73
    cut
    -0.72
    terness
    -0.72
    ittee
    -0.71
    phia
    -0.70
    bow
    -0.70
    POSITIVE LOGITS
     teamwork
    0.86
     religion
    0.86
     these
    0.83
     preserving
    0.83
     intermedi
    0.80
     pronouns
    0.80
     geography
    0.79
     cryptography
    0.78
     informants
    0.76
     textual
    0.76
    Activations Density 0.164%

    No Known Activations