© 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. 3061
    Prev
    Next
    INDEX
    Explanations

    instances of the word "cases" followed by a comma and another word or number

    oai_token-act-pair · gpt-3.5-turbo

    references to specific cases or instances

    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
    agra
    -0.86
    ouch
    -0.74
    1001
    -0.71
    ãĤ´
    -0.71
    mor
    -0.70
    talk
    -0.69
    netflix
    -0.68
    QUI
    -0.67
    ACY
    -0.66
    ãĥ¥
    -0.65
    POSITIVE LOGITS
     applicants
    0.81
     defendants
    0.81
     claimants
    0.75
     attackers
    0.75
     individuals
    0.74
     respondents
    0.74
     involve
    0.73
     assailants
    0.72
     relying
    0.72
     involving
    0.72
    Activations Density 0.206%

    No Known Activations