INDEX

Explanations

words related to summarizing or encapsulating information

oai_token-act-pair · gpt-3.5-turbo

concepts related to summarization and encapsulation

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

eland

-0.75

je

-0.66

TG

-0.66

watch

-0.65

bies

-0.64

emade

-0.61

 Gamble

-0.61

lands

-0.60

bie

-0.58

Je

-0.58

POSITIVE LOGITS

 summar

1.00

 succinct

0.99

 summed

0.94

 artic

0.87

 summarized

0.84

 encaps

0.83

 concise

0.83

initions

0.82

 summarize

0.77

spection

0.76

Activations Density 0.053%

words related to summarizing or encapsulating information

concepts related to summarization and encapsulation

No Comments

No Known Activations