Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 9-RES-JB
    6. 10442
    Prev
    Next
    INDEX
    Explanations

    language indicating something is specially kept aside or not ordinarily noticed or used, such as "reserved for"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "reserved."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MAP
    -0.82
    psc
    -0.70
    ODE
    -0.70
    EEK
    -0.70
    enegger
    -0.69
    sonian
    -0.66
    kok
    -0.65
    HOME
    -0.64
     pus
    -0.63
    akeru
    -0.63
    POSITIVE LOGITS
     reserved
    1.02
    folios
    0.93
     Reserved
    0.89
     judgement
    0.88
     scarce
    0.84
     judgment
    0.81
    ldom
    0.79
     seating
    0.79
     discretion
    0.78
     seats
    0.76
    Activations Density 0.066%

    No Known Activations