INDEX

Explanations

phrases containing the prefix "non-" and related terms

oai_token-act-pair · gpt-3.5-turbo

phrases that denote non-physical characteristics or statuses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Tycoon

-0.94

 Oaks

-0.76

 Spoon

-0.73

 Grill

-0.72

 Grind

-0.72

 Showdown

-0.72

Gry

-0.69

 McGee

-0.68

Dug

-0.67

 Takes

-0.66

POSITIVE LOGITS

chal

1.25

linear

1.22

verbal

1.17

zero

1.14

technical

1.13

fiction

1.12

stop

1.08

profit

1.07

lethal

1.06

etheless

1.06

Activations Density 0.026%

phrases containing the prefix "non-" and related terms

phrases that denote non-physical characteristics or statuses

No Comments

No Known Activations