INDEX

Explanations

places or locations mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

phrases indicating places and their significance or attributes

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

telling

-0.75

 PsyNet

-0.71

ebus

-0.64

ernaut

-0.62

bps

-0.62

hirt

-0.60

 sshd

-0.58

issance

-0.58

cn

-0.58

RN

-0.57

POSITIVE LOGITS

igham

0.67

 habitable

0.67

 Eucl

0.66

 geographically

0.65

 Discrimination

0.63

Interstitial

0.61

 Viper

0.61

 doors

0.61

 vantage

0.61

Vas

0.60

Activations Density 0.454%

places or locations mentioned in the text

phrases indicating places and their significance or attributes

No Comments

No Known Activations