© 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. 5032
    Prev
    Next
    INDEX
    Explanations

    and activate on references or citations within brackets [ ]

    oai_token-act-pair · gpt-3.5-turbo

    references to brackets or bracketed information

    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
     intensive
    -0.70
     maximum
    -0.67
     clash
    -0.65
     srfAttach
    -0.64
     interf
    -0.62
     regression
    -0.62
     embargo
    -0.60
     extinction
    -0.60
     rematch
    -0.60
     cones
    -0.58
    POSITIVE LOGITS
    Pg
    1.27
    âĢ¦]
    1.26
    sic
    1.19
    ...]
    1.17
    ?]
    1.08
    :]
    1.06
    actionDate
    1.05
    .]
    0.93
    paragraph
    0.92
    src
    0.92
    Activations Density 0.036%

    No Known Activations