INDEX

Explanations

words related to laziness and negative judgments

oai_token-act-pair · gpt-3.5-turbo

references to laziness or being lazy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

orial

-0.81

metics

-0.80

endez

-0.75

ellation

-0.74

lished

-0.72

tain

-0.72

iameter

-0.72

esan

-0.71

inction

-0.71

ORN

-0.71

POSITIVE LOGITS

 Bastard

0.96

 lazy

0.90

 jerk

0.82

glers

0.78

 bastard

0.78

 bast

0.77

azy

0.76

sus

0.75

laz

0.72

igue

0.70

Activations Density 0.023%

words related to laziness and negative judgments

references to laziness or being lazy

No Comments

No Known Activations