INDEX

Explanations

phrases related to deception or false statements

oai_token-act-pair · gpt-3.5-turbo

instances of the word "lying," which indicates dishonesty or deceit

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Ultra

-0.86

obs

-0.79

ISO

-0.78

ilation

-0.74

joining

-0.70

FN

-0.70

iles

-0.69

aldi

-0.69

Specific

-0.67

ORE

-0.65

POSITIVE LOGITS

 horizont

0.78

 lying

0.78

lie

0.72

 liar

0.71

utenant

0.71

 pills

0.70

 siege

0.70

 skelet

0.70

acies

0.70

 lied

0.70

Activations Density 0.008%

phrases related to deception or false statements

instances of the word "lying," which indicates dishonesty or deceit

No Comments

No Known Activations