INDEX

Explanations

comparisons or similarities between different things

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Flavoring

-0.98

 Deaths

-0.88

 Answers

-0.76

 VIDEOS

-0.73

 Landing

-0.70

 Ending

-0.70

 Releases

-0.69

 succession

-0.68

psons

-0.68

POSITIVE LOGITS

 comfortable

1.19

 attractive

1.11

 efficient

1.10

 powerful

1.10

 sensitive

1.09

 difficult

1.09

 complicated

1.07

 interesting

1.07

 uncomfortable

1.06

 reliant

1.05

Activations Density 1.556%

comparisons or similarities between different things

No Comments

No Known Activations