INDEX

Explanations

phrases related to deception and concealment

oai_token-act-pair · gpt-3.5-turbo

expressions related to concealing or hiding information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

achi

-0.74

pour

-0.73

iaries

-0.69

bos

-0.68

 neighb

-0.67

psc

-0.66

quad

-0.66

nesota

-0.65

 fitt

-0.64

agues

-0.64

POSITIVE LOGITS

 wrongdoing

1.07

 scandals

0.98

 corruption

0.90

 inconvenient

0.88

 atrocities

0.83

 abuses

0.83

 deception

0.83

 complicity

0.83

 suspicions

0.82

 incompetence

0.82

Activations Density 0.136%

phrases related to deception and concealment

expressions related to concealing or hiding information

No Comments

No Known Activations