INDEX

Explanations

words related to geographical poles

oai_token-act-pair · gpt-3.5-turbo

references to geographic poles or pole-like concepts

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

æĸ¹

-1.00

Interstitial

-0.87

Export

-0.87

esis

-0.78

ebook

-0.78

HR

-0.77

ITNESS

-0.77

iencies

-0.74

ADRA

-0.74

VERTIS

-0.73

POSITIVE LOGITS

 pole

1.43

 poles

1.22

pole

1.12

 Pole

1.02

mic

0.91

 Poles

0.88

 maiden

0.80

ball

0.76

abases

0.75

 dancer

0.73

Activations Density 0.013%

words related to geographical poles

references to geographic poles or pole-like concepts

No Comments

No Known Activations