INDEX

Explanations

words related to maps and geographical locations

oai_token-act-pair · gpt-3.5-turbo

references to maps and mapping-related content

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

nuts

-0.81

nder

-0.76

aneous

-0.73

icer

-0.70

ivas

-0.68

weeney

-0.68

ruciating

-0.68

endiary

-0.67

ndra

-0.66

encers

-0.66

POSITIVE LOGITS

map

0.87

ographer

0.82

map

0.79

 maps

0.77

MAP

0.75

MAP

0.74

sheets

0.72

Map

0.72

makers

0.72

 mapped

0.72

Activations Density 0.019%

words related to maps and geographical locations

references to maps and mapping-related content

No Comments

No Known Activations