INDEX

Explanations

information or mentions related to news articles and reporting

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ilater

-0.58

gall

-0.52

apsed

-0.50

ypes

-0.50

unal

-0.49

andi

-0.47

estern

-0.47

ublished

-0.47

 badge

-0.46

vous

-0.46

POSITIVE LOGITS

 THERE

0.97

 THEY

0.93

there

0.91

 there

0.89

 they

0.79

they

0.71

we

0.69

it

0.68

 There

0.67

WHY

0.67

Activations Density 0.728%

information or mentions related to news articles and reporting

No Comments

No Known Activations