INDEX

Explanations

mentions of the color gray

oai_token-act-pair · gpt-3.5-turbo

references to the color gray

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

=-=-=-=-

-0.85

CVE

-0.83

ugal

-0.78

========

-0.78

ÄŁ

-0.77

olkien

-0.76

andals

-0.74

igslist

-0.73

Ì¶

-0.72

ãĥ¼ãĥ³

-0.72

POSITIVE LOGITS

hound

1.01

beard

0.94

 shading

0.91

haired

0.86

ish

0.85

 gray

0.84

 grey

0.80

 linen

0.79

naire

0.77

 slime

0.77

Activations Density 0.022%

mentions of the color gray

references to the color gray

No Comments

No Known Activations