INDEX

Explanations

phrases related to exploring and navigating different locations

oai_token-act-pair · gpt-3.5-turbo

references to exploration and travel within 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

ufact

-0.71

erity

-0.67

endment

-0.65

thood

-0.64

elected

-0.63

ajor

-0.62

rss

-0.62

ANCE

-0.61

ACP

-0.60

-+-+-+-+

-0.59

POSITIVE LOGITS

 labyrinth

1.17

 maze

1.06

 depths

0.98

 corridors

0.96

yrinth

0.91

 halls

0.90

 boundaries

0.90

 gardens

0.86

 tunnels

0.85

 realms

0.85

Activations Density 0.383%

phrases related to exploring and navigating different locations

references to exploration and travel within various contexts

No Comments

No Known Activations