INDEX

Explanations

terms related to breathing or respiratory issues

oai_token-act-pair · gpt-3.5-turbo

variations of the word "treat" indicating different contexts or usages

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

 Zimbabwe

-0.68

PLIC

-0.67

 Tanzania

-0.64

OTA

-0.63

BIL

-0.63

BOX

-0.63

 siph

-0.62

 Scotia

-0.60

 Laurel

-0.59

 deleting

-0.58

POSITIVE LOGITS

htaking

1.22

idon

1.04

ty

1.02

iful

1.01

eous

0.98

hing

0.98

hs

0.97

hes

0.95

sburg

0.94

terson

0.94

Activations Density 0.023%

terms related to breathing or respiratory issues

variations of the word "treat" indicating different contexts or usages

No Comments

No Known Activations