INDEX

Explanations

words related to discussions or discourses

oai_token-act-pair · gpt-3.5-turbo

instances of the word "discriminate" and its variations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

mberg

-0.78

cki

-0.76

tti

-0.73

Interstitial

-0.68

akeru

-0.66

 deed

-0.65

ettings

-0.64

 grace

-0.63

Downloadha

-0.62

VILLE

-0.61

POSITIVE LOGITS

ussions

1.25

overed

1.17

ourse

1.12

ount

1.09

ontent

1.07

ussed

1.07

iple

1.06

ussion

1.06

ourses

1.02

retion

0.98

Activations Density 0.019%

words related to discussions or discourses

instances of the word "discriminate" and its variations

No Comments

No Known Activations