© 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. 0-RES-JB
    11. 4034
    Prev
    Next
    INDEX
    Explanations

    phrases related to a person named Adrian

    oai_token-act-pair · gpt-3.5-turbo

    mentions of people named Adrian

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    CLOSE
    -0.74
    OTUS
    -0.70
    7601
    -0.67
     fail
    -0.66
     cond
    -0.64
    eq
    -0.62
     fulfillment
    -0.62
    psc
    -0.62
    .>>
    -0.60
     failure
    -0.59
    POSITIVE LOGITS
     Adrian
    3.52
     Emil
    1.36
     Darren
    1.36
     Damian
    1.33
     Gian
    1.30
     Damien
    1.24
     Ian
    1.23
     Julian
    1.22
     Petra
    1.21
     Olivier
    1.18
    Activations Density 0.017%

    No Known Activations