INDEX

Explanations

mentions of chemical weapons-related terms and events

oai_token-act-pair · gpt-3.5-turbo

references to chemical weapons and their usage

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

alg

-0.88

unct

-0.86

ATA

-0.81

 Huck

-0.73

erenn

-0.73

uckland

-0.72

Stud

-0.72

aird

-0.72

YC

-0.71

osit

-0.71

POSITIVE LOGITS

 bombing

1.20

 bombings

1.11

 munitions

1.09

 disarm

1.03

 bombs

1.03

 weapons

1.00

 grenades

0.99

 airstrikes

0.98

apons

0.95

 attacks

0.94

Activations Density 0.060%

mentions of chemical weapons-related terms and events

references to chemical weapons and their usage

No Comments

No Known Activations