© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 10-RES-JB
    11. 17474
    Prev
    Next
    INDEX
    Explanations

    phrases indicating reclaiming or retrieval

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to actions of returning or bringing something back

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    achi
    -0.99
     Ich
    -0.94
    ECT
    -0.92
     TAMADRA
    -0.89
    CE
    -0.88
    Sense
    -0.84
     Phill
    -0.83
    Phill
    -0.82
     Collins
    -0.78
     Phillips
    -0.78
    POSITIVE LOGITS
    BACK
    1.17
     BACK
    1.16
    back
    1.14
     backs
    1.13
     back
    1.08
     Back
    1.03
    backs
    1.01
     backed
    0.97
     backward
    0.97
     backwards
    0.94
    Activations Density 0.239%

    No Known Activations