INDEX

Explanations

political terms and phrases

oai_token-act-pair · gpt-3.5-turbo

terms related to political and social concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

aneers

-0.80

©¶æ

-0.78

ahs

-0.73

»Ĵ

-0.72

tyard

-0.71

sers

-0.68

onds

-0.68

icrobial

-0.67

mosp

-0.66

racuse

-0.66

POSITIVE LOGITS

 enough

1.17

 territory

0.95

 fodder

0.89

 insofar

0.87

 folly

0.85

 nonsense

0.85

aneously

0.85

 because

0.84

ALLY

0.84

0.83

Activations Density 0.374%

political terms and phrases

terms related to political and social concepts

No Comments

No Known Activations