INDEX

Explanations

words related to undercover operations

oai_token-act-pair · gpt-3.5-turbo

mentions of undercover operations and agents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

gran

-0.74

itsch

-0.72

grave

-0.71

È

-0.70

cing

-0.68

 Forge

-0.67

Pwr

-0.67

ogun

-0.66

RR

-0.66

akeru

-0.66

POSITIVE LOGITS

 undercover

1.16

 informant

1.06

 informants

0.87

 sting

0.86

 infiltr

0.85

 mole

0.84

 detective

0.76

 operatives

0.76

 investigator

0.76

 taped

0.75

Activations Density 0.014%

words related to undercover operations

mentions of undercover operations and agents

No Comments

No Known Activations