INDEX

Explanations

phrases related to evaluation or judgment

oai_token-act-pair · gpt-3.5-turbo

articles that indicate the significance or importance of a topic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Edit

-0.94

books

-0.81

 edits

-0.73

Versions

-0.70

 airs

-0.70

edit

-0.67

writers

-0.66

alks

-0.65

Events

-0.65

 Adin

-0.64

POSITIVE LOGITS

 reminder

1.14

lot

1.08

 prerequisite

1.08

 testament

1.08

bit

1.04

 reflection

1.00

rouse

1.00

versive

1.00

 boon

1.00

 hallmark

0.99

Activations Density 0.277%

phrases related to evaluation or judgment

articles that indicate the significance or importance of a topic

No Comments

No Known Activations