Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 14011
    Prev
    Next
    INDEX
    Explanations

    questions about events and consequences

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    14
    +0.16
    0.5%
    542
    +0.14
    0.5%
    718
    +0.14
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    542
    +0.16
    0.04
    588
    +0.14
    0.04
    20
    +0.14
    0.04
    Negative Logits
    ament
    -1.25
    tesy
    -1.25
    cius
    -1.21
    mens
    -1.17
    staff
    -1.16
    bors
    -1.13
    overed
    -1.11
    recy
    -1.07
    rehend
    -1.07
    ways
    -1.06
    POSITIVE LOGITS
    uate
    1.28
     unfold
    1.27
     happen
    1.19
     unfolding
    1.19
    uates
    1.14
     happening
    1.07
     havoc
    1.03
    uating
    0.99
     transpired
    0.97
     happens
    0.97
    Activations Density 0.359%

    No Known Activations