INDEX

Explanations

references to terrorist activities

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

HUD

-0.73

è£

-0.68

DOC

-0.67

 Reflect

-0.63

 Bench

-0.62

natureconservancy

-0.62

oqu

-0.60

 offic

-0.60

uci

-0.59

 consulted

-0.59

POSITIVE LOGITS

 unsuspecting

1.01

 havoc

1.01

 undet

1.01

 revenge

0.97

 sleeper

0.94

 foothold

0.90

 terror

0.89

 infect

0.87

 mayhem

0.85

 infiltr

0.85

Activations Density 0.402%

references to terrorist activities

No Comments

No Known Activations