INDEX

Explanations

specific mentions of speeches and addresses

oai_token-act-pair · gpt-3.5-turbo

phrases related to speeches or addresses made to organizations or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

usercontent

-0.83

nah

-0.80

 workaround

-0.76

 ILCS

-0.72

chrome

-0.68

bear

-0.68

 discovered

-0.68

 droid

-0.66

afe

-0.65

taboola

-0.64

POSITIVE LOGITS

 keynote

1.00

 televised

0.94

 podium

0.90

 speeches

0.90

 commencement

0.89

 rallies

0.87

ä¹ĭ

0.86

 rally

0.81

interstitial

0.81

 seminars

0.80

Activations Density 0.279%

specific mentions of speeches and addresses

phrases related to speeches or addresses made to organizations or groups

No Comments

No Known Activations