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. 8515
    Prev
    Next
    INDEX
    Explanations

    descriptions of biological structures and processes

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of biological structures and their characteristics

    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
    gery
    -0.87
     Armageddon
    -0.75
     Facts
    -0.69
    idon
    -0.67
    shit
    -0.66
     Setup
    -0.65
    chnology
    -0.65
    akery
    -0.65
     GOT
    -0.63
    got
    -0.62
    POSITIVE LOGITS
     contiguous
    1.21
     spaced
    1.18
     narrower
    1.17
     clustered
    1.12
     dispersed
    1.10
     indist
    1.09
     irregular
    1.09
     densely
    1.08
     perpendicular
    1.07
     shorter
    1.07
    Activations Density 0.429%

    No Known Activations