INDEX

Explanations

instances where things are being utilized or employed

oai_token-act-pair · gpt-3.5-turbo

instances of the word "used" in various contexts

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

Hai

-0.72

 Tale

-0.71

gdala

-0.68

nesty

-0.66

mare

-0.65

awk

-0.65

COMPLE

-0.63

odore

-0.62

dom

-0.61

bitious

-0.61

POSITIVE LOGITS

 sparing

0.91

fully

0.83

 Used

0.78

 extensively

0.72

FUL

0.72

 interchange

0.70

 condoms

0.68

MSN

0.68

 wielded

0.67

 wisely

0.66

Activations Density 0.053%

instances where things are being utilized or employed

instances of the word "used" in various contexts

No Comments

No Known Activations