© 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. 185
    Prev
    Next
    INDEX
    Explanations

    phrases related to breaking in or breaking out of places

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to breaking in or breaking out actions

    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
     til
    -0.73
     postp
    -0.72
     Trip
    -0.65
    forward
    -0.63
     priorit
    -0.60
     overs
    -0.59
    ãĥ¼ãĤ¯
    -0.57
     alike
    -0.57
     dodged
    -0.57
    lists
    -0.56
    POSITIVE LOGITS
    ydia
    0.71
    ymes
    0.69
    ochet
    0.69
    erker
    0.67
    taining
    0.65
     cloves
    0.65
     spontaneously
    0.64
    ody
    0.63
     tears
    0.61
    tain
    0.61
    Activations Density 0.070%

    No Known Activations