INDEX

Explanations

keywords related to hazmat or hazardous materials

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Hazard."

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

utsche

-0.77

ļéĨĴ

-0.77

ngth

-0.73

istic

-0.73

ounter

-0.72

intendo

-0.72

ãĥ¤

-0.71

izes

-0.68

ãĤ´ãĥ³

-0.68

IZE

-0.67

POSITIVE LOGITS

Haz

1.09

ards

1.04

arding

1.04

bolt

0.83

eas

0.83

arro

0.81

ARDS

0.80

mat

0.80

stack

0.79

engers

0.78

Activations Density 0.012%

keywords related to hazmat or hazardous materials

mentions of the word "Hazard."

No Comments

No Known Activations