INDEX

Explanations

references to physical structures like wings

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "wing" or related terms

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

ãĥ¤

-0.74

Uz

-0.70

 metab

-0.70

Interstitial

-0.69

 Cumber

-0.67

 Glock

-0.64

 IMAGES

-0.64

 Nare

-0.63

CBI

-0.63

âĶĢâĶĢ

-0.62

POSITIVE LOGITS

tips

1.38

tip

1.36

nuts

1.32

nut

1.23

edly

1.03

 Chun

0.93

erness

0.93

ners

0.91

notes

0.89

bill

0.89

Activations Density 0.019%

references to physical structures like wings

occurrences of the word "wing" or related terms

No Comments

No Known Activations