INDEX

Explanations

years from the 1990s

oai_token-act-pair · gpt-3.5-turbo

references to the year 1990 and its related events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

holder

-0.75

bow

-0.71

nery

-0.70

cho

-0.67

hire

-0.66

pta

-0.66

 hungry

-0.65

semble

-0.65

holders

-0.65

cloth

-0.65

POSITIVE LOGITS

ĸļ

1.21

sburg

0.75

ãĤ¦ãĤ¹

0.74

cture

0.74

 GOODMAN

0.70

ãĥŁ

0.70

å¹

0.69

-'

0.68

 ï¿½ï¿½ï¿½ï¿½

0.67

ãĥ³

0.66

Activations Density 0.017%

years from the 1990s

references to the year 1990 and its related events

No Comments

No Known Activations