INDEX

Explanations

quantitative data related to different categories like numbers and percentages

oai_token-act-pair · gpt-3.5-turbo

numeric values associated with statistical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 smugg

-0.69

 ILCS

-0.63

icans

-0.62

meet

-0.61

 Canaver

-0.61

 tailor

-0.60

 wardrobe

-0.59

 hiber

-0.58

 dressing

-0.56

SpaceEngineers

-0.56

POSITIVE LOGITS

1.11

 whopping

1.05

1.02

1.01

0.99

0.98

0.97

Activations Density 0.433%

quantitative data related to different categories like numbers and percentages

numeric values associated with statistical data

No Comments

No Known Activations