INDEX

Explanations

phrases related to using a specific tool or method

oai_token-act-pair · gpt-3.5-turbo

instances of the word "using."

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

owship

-0.76

uberty

-0.76

hound

-0.74

nesty

-0.72

witz

-0.70

brother

-0.70

spawn

-0.65

stal

-0.65

hood

-0.65

cffffcc

-0.64

POSITIVE LOGITS

FUL

0.83

 borrowed

0.74

 contacts

0.70

 computers

0.70

iences

0.68

fully

0.68

Offline

0.67

 shortcuts

0.66

 techniques

0.65

 sparing

0.65

Activations Density 0.045%

phrases related to using a specific tool or method

instances of the word "using."

No Comments

No Known Activations