© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 13632
    Prev
    Next
    INDEX
    Explanations

    encoded text containing a mix of different languages

    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
    647
    +0.16
    0.6%
    167
    +0.16
    0.5%
    730
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    176
    +0.16
    0.03
    229
    +0.16
    0.03
    647
    +0.15
    0.03
    Negative Logits
    oulos
    -1.16
    wagen
    -1.11
    theless
    -0.95
    creen
    -0.90
    terday
    -0.90
    osate
    -0.88
     compromise
    -0.88
    obyl
    -0.87
    ffield
    -0.87
     guiActiveUnfocused
    -0.86
    POSITIVE LOGITS
    ľ
    1.25
    ĭ
    1.19
    Ĥ
    1.18
    ¦
    1.16
    ģ
    1.13
    irl
    1.12
    urn
    1.11
    ł
    1.10
    irc
    1.10
    urch
    1.09
    Activations Density 0.154%

    No Known Activations