INDEX

Explanations

medical conditions and drug-related information correlated with numbers or categories

oai_token-act-pair · gpt-3.5-turbo

references to medical conditions and classifications

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

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

ãĥĨãĤ£

-0.86

owe

-0.70

isman

-0.68

Week

-0.67

Sports

-0.67

mith

-0.66

rosc

-0.65

netflix

-0.65

nick

-0.64

follow

-0.63

POSITIVE LOGITS

 inhibitor

0.87

 inhibitors

0.85

 guiActiveUnfocused

0.78

 Intermediate

0.74

 attackers

0.66

²¾

0.65

 Centauri

0.65

 defenses

0.64

 booster

0.63

xual

0.63

Activations Density 0.087%

medical conditions and drug-related information correlated with numbers or categories

references to medical conditions and classifications

No Comments

No Known Activations