INDEX

Explanations

words related to the location "Australia"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the token "aus"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 CARD

-0.68

EVA

-0.65

Assembly

-0.65

BOOK

-0.65

JECT

-0.64

TL

-0.62

dfx

-0.61

 short

-0.61

 Sprite

-0.60

 lear

-0.59

POSITIVE LOGITS

aus

1.28

chwitz

1.05

uin

0.94

inite

0.94

ilon

0.91

uve

0.90

hent

0.90

nown

0.90

ername

0.86

ertodd

0.86

Activations Density 0.005%

words related to the location "Australia"

occurrences of the token "aus"

No Comments

No Known Activations