© 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. 0-RES-JB
    11. 16541
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences containing the word "middle."

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Shots
    -0.80
    craft
    -0.70
    breakers
    -0.66
     aboard
    -0.66
    hunt
    -0.64
    beam
    -0.63
    soever
    -0.62
     Refer
    -0.62
     Recall
    -0.62
    edit
    -0.62
    POSITIVE LOGITS
     middle
    3.70
    middle
    2.88
    Middle
    2.44
     Middle
    1.83
     midst
    1.44
     MID
    1.29
     mid
    1.28
     intermediate
    1.24
     mids
    1.21
     center
    1.17
    Activations Density 0.024%

    No Known Activations