© 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. 4-RES-JB
    11. 8498
    Prev
    Next
    INDEX
    Explanations

    software or technology products with the term "Pro" in their names

    oai_token-act-pair · gpt-3.5-turbo

    references to "Pro" products or versions, particularly in technology-related contexts

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

    No Comments

    Negative Logits
     halls
    -0.81
    ãĤ¼ãĤ¦ãĤ¹
    -0.79
    owship
    -0.78
     Cornell
    -0.71
     Dise
    -0.67
     straw
    -0.67
    aved
    -0.66
    ablishment
    -0.65
     spears
    -0.64
     needles
    -0.64
    POSITIVE LOGITS
    digy
    1.48
    posal
    1.24
    secution
    1.18
    gression
    1.18
    portion
    1.13
    verbs
    1.13
    poses
    1.10
    pose
    1.08
    hibited
    1.05
    blems
    1.03
    Activations Density 0.010%

    No Known Activations