INDEX

Explanations

words related to deception and falsehoods

oai_token-act-pair · gpt-3.5-turbo

terms related to dishonesty and deception

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

airo

-0.72

ateur

-0.71

imen

-0.71

cise

-0.70

aldo

-0.67

enhagen

-0.65

 Alto

-0.65

aird

-0.64

foreseen

-0.64

rises

-0.63

POSITIVE LOGITS

 perpetrated

0.98

 detector

0.92

ulent

0.89

ulence

0.82

 pedd

0.81

 falsely

0.79

telling

0.76

 omission

0.76

uten

0.75

ument

0.74

Activations Density 0.088%

words related to deception and falsehoods

terms related to dishonesty and deception

No Comments

No Known Activations