© 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. 4-RES-JB
    11. 14918
    Prev
    Next
    INDEX
    Explanations

    references to legal proceedings and legal professionals

    oai_token-act-pair · gpt-3.5-turbo

    references to legal professionals and regulatory bodies

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

    No Comments

    Negative Logits
    nings
    -0.61
     coma
    -0.53
     metaphor
    -0.52
     prol
    -0.52
     comeback
    -0.51
     sunset
    -0.51
     loneliness
    -0.51
     arcade
    -0.50
     Ashes
    -0.50
    terness
    -0.49
    POSITIVE LOGITS
     tasked
    0.63
     instructed
    0.62
    ij士
    0.62
     overseeing
    0.62
     oversees
    0.60
     convened
    0.60
     ABE
    0.60
     specializing
    0.60
    ngth
    0.59
     oversee
    0.59
    Activations Density 2.063%

    No Known Activations