© 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. 10-RES-JB
    11. 14078
    Prev
    Next
    INDEX
    Explanations

    keywords related to financial investments and program upgrades, especially phrases referring to local contractors and percentage inclusion

    oai_token-act-pair · gpt-3.5-turbo

    instances of uncertainty or unexpected outcomes

    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
     thereto
    -0.66
     to
    -0.66
     ta
    -0.55
     To
    -0.54
     unto
    -0.54
    to
    -0.53
     TO
    -0.52
     edits
    -0.52
     stances
    -0.51
    To
    -0.51
    POSITIVE LOGITS
    bably
    0.64
    prisingly
    0.62
    yssey
    0.59
    Newsletter
    0.59
    usual
    0.57
    except
    0.57
     hope
    0.56
    named
    0.56
    ocent
    0.56
     expecting
    0.54
    Activations Density 0.633%

    No Known Activations