INDEX

Explanations

terms related to specific individuals, organizations, or events portrayed in various contexts

oai_token-act-pair · gpt-3.5-turbo

key terms and phrases related to specific professions and roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

bows

-0.86

Apps

-0.73

 iPads

-0.73

 Governments

-0.72

 kitchens

-0.70

 avenues

-0.68

rooms

-0.66

 Parables

-0.65

 Sections

-0.65

 testimonies

-0.64

POSITIVE LOGITS

 whose

0.96

who

0.85

 whom

0.83

hander

0.83

 clone

0.83

 jumper

0.80

whose

0.79

 named

0.79

 resembling

0.78

 racer

0.78

Activations Density 0.605%

terms related to specific individuals, organizations, or events portrayed in various contexts

key terms and phrases related to specific professions and roles

No Comments

No Known Activations