INDEX

Explanations

references to the city of Pasadena

oai_token-act-pair · gpt-3.5-turbo

references to specific geographical locations or place names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

plates

-0.79

onut

-0.78

appings

-0.77

ycle

-0.77

hya

-0.76

ivities

-0.75

ikuman

-0.73

doms

-0.72

lass

-0.72

phas

-0.72

POSITIVE LOGITS

 Pasadena

0.98

 Granger

0.94

xual

0.84

 Superior

0.77

 Observatory

0.76

xon

0.76

adena

0.75

 Parenthood

0.72

 Coyotes

0.72

 Advent

0.69

Activations Density 0.017%

references to the city of Pasadena

references to specific geographical locations or place names

No Comments

No Known Activations