INDEX

Explanations

positive adjectives describing attributes or qualities

oai_token-act-pair · gpt-3.5-turbo

adjectives that describe intensity or quality

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

 helicop

-0.75

ilaterally

-0.64

farious

-0.63

adata

-0.62

artney

-0.62

pheus

-0.62

anamo

-0.62

addafi

-0.59

stood

-0.59

aea

-0.57

POSITIVE LOGITS

 Disclosure

0.60

CRIPTION

0.53

 unfocusedRange

0.52

SVG

0.51

 Posted

0.51

0.50

 mango

0.50

actionDate

0.49

 nutshell

0.49

 crochet

0.48

Activations Density 0.333%

positive adjectives describing attributes or qualities

adjectives that describe intensity or quality

No Comments

No Known Activations