INDEX

Explanations

phrases related to legal or governmental processes

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

externalActionCode

-0.81

been

-0.75

yond

-0.71

é¾įå¥ĳå£«

-0.71

wcsstore

-0.69

é¾įåĸļå£«

-0.68

isable

-0.67

omorphic

-0.67

lie

-0.66

 anymore

-0.65

POSITIVE LOGITS

Wem

0.79

 extensively

0.76

eatures

0.74

 yesterday

0.69

nesday

0.69

 Horowitz

0.68

 unsuccessfully

0.68

 Levine

0.68

its

0.66

 Greenberg

0.66

Activations Density 0.371%

phrases related to legal or governmental processes

No Comments

No Known Activations