© 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. 8-RES-JB
    11. 6207
    Prev
    Next
    INDEX
    Explanations

    phrases starting with quotations

    oai_token-act-pair · gpt-3.5-turbo

    instances of direct speech

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     princip
    -0.85
     flared
    -0.81
     deterrent
    -0.77
     bod
    -0.72
     cheek
    -0.72
     clut
    -0.72
     arri
    -0.71
     conve
    -0.71
     adv
    -0.70
     sund
    -0.70
    POSITIVE LOGITS
    Hey
    1.54
    hey
    1.41
    Oh
    1.39
    hello
    1.34
    Damn
    1.25
    why
    1.25
    Fuck
    1.25
    Look
    1.24
    Why
    1.23
    Okay
    1.23
    Activations Density 0.049%

    No Known Activations