INDEX

Explanations

specific car models with mention of their year of origin

oai_token-act-pair · gpt-3.5-turbo

names or labels related to various concepts, likely categorizing or indexing items

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

awar

-0.76

wcs

-0.73

turned

-0.70

better

-0.69

ordering

-0.68

ighth

-0.66

comments

-0.66

perty

-0.65

appropriately

-0.64

gewater

-0.64

POSITIVE LOGITS

 Presents

0.79

 Industries

0.78

 Devices

0.77

 Enterprises

0.77

geist

0.75

 Technician

0.75

 Madness

0.75

 Hebdo

0.74

 Videos

0.73

 Productions

0.73

Activations Density 0.199%

specific car models with mention of their year of origin

names or labels related to various concepts, likely categorizing or indexing items

No Comments

No Known Activations