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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 11-RES-JB
    6. 9533
    Prev
    Next
    INDEX
    Explanations

    phrases mentioning exclusivity or limited numbers of people

    oai_token-act-pair · gpt-3.5-turbo

    terms related to exclusivity and limited access

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    moil
    -0.63
     nowhere
    -0.63
    rus
    -0.60
    çľ
    -0.59
    auga
    -0.58
    apego
    -0.58
    nothing
    -0.58
     Translation
    -0.58
     DRAG
    -0.57
    çī
    -0.57
    POSITIVE LOGITS
     survive
    1.29
     survives
    1.22
     ever
    1.17
    ever
    1.15
     anymore
    1.11
     survived
    1.07
     realise
    1.04
     bothered
    1.01
     dared
    1.01
     actually
    0.99
    Activations Density 0.224%

    No Known Activations