© 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. 1-RES-JB
    11. 874
    Prev
    Next
    INDEX
    Explanations

    instances of makeshift or impromptu solutions or setups

    oai_token-act-pair · gpt-3.5-turbo

    terms related to temporary or makeshift constructions and improvisation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    speed
    -0.80
    thood
    -0.77
    cross
    -0.75
    hips
    -0.74
     Thro
    -0.73
    alties
    -0.71
    OA
    -0.70
    worthiness
    -0.69
    rich
    -0.69
    inion
    -0.69
    POSITIVE LOGITS
     makeshift
    1.18
     encamp
    0.95
     improvised
    0.90
    romptu
    0.83
     memorial
    0.83
     tent
    0.81
     benches
    0.79
     roadside
    0.77
    LV
    0.76
     stret
    0.76
    Activations Density 0.015%

    No Known Activations