INDEX

Explanations

keywords related to places or things like "garage."

oai_token-act-pair · gpt-3.5-turbo

references to physical locations and structures

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

 Tribe

-0.79

PORT

-0.78

DragonMagazine

-0.78

BSD

-0.77

ãĤ¼

-0.75

DEV

-0.75

CV

-0.74

ãĥ´ãĤ¡

-0.72

 Club

-0.71

Revolution

-0.70

POSITIVE LOGITS

oling

0.97

inged

0.91

 todd

0.87

oled

0.86

aded

0.84

gue

0.84

iper

0.84

icol

0.81

 jung

0.80

etime

0.80

Activations Density 0.047%

keywords related to places or things like "garage."

references to physical locations and structures

No Comments

No Known Activations