INDEX

Explanations

words related to sharp objects, specifically knives

oai_token-act-pair · gpt-3.5-turbo

references to knives in various contexts

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

mberg

-0.86

rian

-0.82

rians

-0.81

oday

-0.78

tainment

-0.78

rium

-0.75

ICLE

-0.74

ALS

-0.72

rix

-0.71

 Emin

-0.71

POSITIVE LOGITS

 knives

1.05

 knife

1.02

knife

1.00

 blades

0.99

 scissors

0.98

 blade

0.97

 slicing

0.91

 cutter

0.87

 wielding

0.87

 claws

0.86

Activations Density 0.025%

words related to sharp objects, specifically knives

references to knives in various contexts

No Comments

No Known Activations