INDEX

Explanations

the word "just" used in different contexts

oai_token-act-pair · gpt-3.5-turbo

the word "just" and its various contexts in the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

luaj

-0.72

yrinth

-0.72

 seiz

-0.72

thens

-0.70

then

-0.70

 Also

-0.69

SEE

-0.69

apolis

-0.69

keyes

-0.69

pora

-0.68

POSITIVE LOGITS

ifiable

1.49

 plain

1.05

 icing

0.97

ifications

0.97

 kidding

0.90

 another

0.86

 semantics

0.84

ified

0.84

 decoration

0.83

 scratching

0.82

Activations Density 0.047%

the word "just" used in different contexts

the word "just" and its various contexts in the document

No Comments

No Known Activations