INDEX

Explanations

opponents or opposition to a specific topic or concept

oai_token-act-pair · gpt-3.5-turbo

references to opponents or adversaries in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Peninsula

-0.80

 peninsula

-0.72

 Loch

-0.71

red

-0.69

LOC

-0.68

val

-0.68

 Forestry

-0.67

vol

-0.66

vest

-0.65

ahon

-0.64

POSITIVE LOGITS

 opponents

1.28

 opponent

1.10

 adversaries

1.06

 foes

1.06

foe

1.02

 strugg

0.98

 challengers

0.95

 batters

0.93

 hitters

0.88

 rivals

0.85

Activations Density 0.008%

opponents or opposition to a specific topic or concept

references to opponents or adversaries in various contexts

No Comments

No Known Activations