© 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. 4814
    Prev
    Next
    INDEX
    Explanations

    words related to the concept of shortening or making something shorter

    oai_token-act-pair · gpt-3.5-turbo

    terms related to reducing length or duration

    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
    Detect
    -0.78
    Fund
    -0.77
    Mass
    -0.75
    Activ
    -0.75
    Gener
    -0.73
    DW
    -0.72
    ILLE
    -0.71
    DC
    -0.70
    Ro
    -0.70
    IDA
    -0.70
    POSITIVE LOGITS
     shortened
    1.11
     shortest
    1.10
     shorter
    1.03
     shorten
    0.98
    short
    0.93
    ftime
    0.90
    thens
    0.86
     commute
    0.85
     leng
    0.85
     postp
    0.83
    Activations Density 0.008%

    No Known Activations