INDEX

Explanations

medical terms related to head injuries and conditions like concussions

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

population

-0.74

Adds

-0.73

abet

-0.72

ells

-0.70

GBT

-0.70

Dial

-0.67

ynt

-0.67

abor

-0.64

ager

-0.62

%"

-0.62

POSITIVE LOGITS

 concussion

0.92

 headache

0.90

 headaches

0.86

 grenades

0.85

 goggles

0.73

 headset

0.73

 migraine

0.71

wcs

0.70

 grenade

0.68

 aver

0.67

Activations Density 0.011%

medical terms related to head injuries and conditions like concussions

No Comments

No Known Activations