INDEX

Explanations

numbers and percentages

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "fe"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 GEAR

-0.75

worthy

-0.73

 ANGEL

-0.72

pora

-0.70

agically

-0.66

nas

-0.66

antically

-0.63

mag

-0.63

azines

-0.63

inates

-0.63

POSITIVE LOGITS

eling

1.15

ck

1.04

cker

1.00

els

0.99

lda

0.98

encing

0.93

elin

0.93

cking

0.92

ffer

0.92

zza

0.91

Activations Density 0.011%

numbers and percentages

occurrences of the substring "fe"

No Comments

No Known Activations